From 76fad3edd9e8a1e68c84ba1fa34ccb3078baf8e6 Mon Sep 17 00:00:00 2001 From: Garry Tan Date: Tue, 17 Nov 2026 17:00:55 +0800 Subject: [PATCH] Make gstack agents lead every report with what failed, what's unverified and what else changed (GST-1455) (#4273) * Make gstack agents lead every report with what failed, what's unverified and what else changed (GST-1455) The shared closer asks a status report to stay under 250 words and open with Failed or skipped:, Unverified: and Also changed: lines, each naming every item or none; lists past 10 give the top 10 and a count. Capy's factory.md point 10, the lane report Outcome and READY's lane lines carry the same lines, and point 10 points findings write-ups at ETHOS.md section 8. The wording is GEVA-2's round-3 candidate f83eba4, ported without the retired /investigate rows. factory.md is 4,396 B, so its budget rises from 4,200 to 4,400. Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> * Update the ship skill golden files for the new shared preamble id (GST-1455) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> --------- Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> --- autoplan/SKILL.md | 2 +- autoplan/sections/preamble.md | 7 +++---- canary/SKILL.md | 2 +- canary/sections/preamble.md | 7 +++---- captain/SKILL.md | 2 +- captain/sections/preamble.md | 7 +++---- captain/templates/READY.md | 2 ++ captain/templates/report.md | 2 +- capy/skill/gstack/factory.md | 2 +- changelog.d/gstack-reports-lead-with-failures.md | 1 + codex/SKILL.md | 2 +- codex/sections/preamble.md | 7 +++---- context-restore/SKILL.md | 2 +- context-restore/sections/preamble.md | 7 +++---- context-save/SKILL.md | 2 +- context-save/sections/preamble.md | 7 +++---- design-consultation/SKILL.md | 2 +- design-consultation/sections/preamble.md | 7 +++---- design-html/SKILL.md | 2 +- design-html/sections/preamble.md | 7 +++---- design-motion/SKILL.md | 2 +- design-motion/sections/preamble.md | 7 +++---- design-review/SKILL.md | 2 +- design-review/sections/preamble.md | 7 +++---- design-shotgun/SKILL.md | 2 +- design-shotgun/sections/preamble.md | 7 +++---- devex-review/SKILL.md | 2 +- devex-review/sections/preamble.md | 7 +++---- document-generate/SKILL.md | 2 +- document-generate/sections/preamble.md | 7 +++---- document-release/SKILL.md | 2 +- document-release/sections/preamble.md | 7 +++---- eval-plan/SKILL.md | 2 +- eval-plan/sections/preamble.md | 7 +++---- health/SKILL.md | 2 +- health/sections/preamble.md | 7 +++---- ios-clean/SKILL.md | 2 +- ios-clean/sections/preamble.md | 7 +++---- ios-design-review/SKILL.md | 2 +- ios-design-review/sections/preamble.md | 7 +++---- ios-fix/SKILL.md | 2 +- ios-fix/sections/preamble.md | 7 +++---- ios-qa/SKILL.md | 2 +- ios-qa/sections/preamble.md | 7 +++---- ios-sync/SKILL.md | 2 +- ios-sync/sections/preamble.md | 7 +++---- land-and-deploy/SKILL.md | 2 +- land-and-deploy/sections/preamble.md | 7 +++---- landing-report/SKILL.md | 2 +- landing-report/sections/preamble.md | 7 +++---- learn/SKILL.md | 2 +- learn/sections/preamble.md | 7 +++---- modernize/SKILL.md | 2 +- modernize/sections/preamble.md | 7 +++---- office-hours/SKILL.md | 2 +- office-hours/sections/preamble.md | 7 +++---- pair-agent/SKILL.md | 2 +- pair-agent/sections/preamble.md | 7 +++---- plan-ceo-review/SKILL.md | 2 +- plan-ceo-review/sections/preamble.md | 7 +++---- plan-design-review/SKILL.md | 2 +- plan-design-review/sections/preamble.md | 7 +++---- plan-devex-review/SKILL.md | 2 +- plan-devex-review/sections/preamble.md | 7 +++---- plan-eng-review/SKILL.md | 2 +- plan-eng-review/sections/preamble.md | 7 +++---- plan-tune/SKILL.md | 2 +- plan-tune/sections/preamble.md | 7 +++---- qa-only/SKILL.md | 2 +- qa-only/sections/preamble.md | 7 +++---- qa/SKILL.md | 2 +- qa/sections/preamble.md | 7 +++---- retro/SKILL.md | 2 +- retro/sections/preamble.md | 7 +++---- review/SKILL.md | 2 +- review/sections/preamble.md | 7 +++---- scripts/resolvers/preamble/generate-voice-directive.ts | 5 ++--- setup-deploy/SKILL.md | 2 +- setup-deploy/sections/preamble.md | 7 +++---- setup-gbrain/SKILL.md | 2 +- setup-gbrain/sections/preamble.md | 7 +++---- ship/SKILL.md | 2 +- ship/sections/preamble.md | 7 +++---- skillify/SKILL.md | 2 +- skillify/sections/preamble.md | 7 +++---- spec/SKILL.md | 2 +- spec/sections/preamble.md | 7 +++---- sync-gbrain/SKILL.md | 2 +- sync-gbrain/sections/preamble.md | 7 +++---- test-audit/SKILL.md | 2 +- test-audit/sections/preamble.md | 7 +++---- test/capy-skill-wrapper.test.ts | 6 ++++-- test/fixtures/golden/agy-ship-SKILL.md | 2 +- test/fixtures/golden/claude-ship-SKILL.md | 2 +- test/fixtures/golden/codex-ship-SKILL.md | 2 +- test/fixtures/golden/factory-ship-SKILL.md | 2 +- .../ship/gstack-reports-lead-with-failures.json | 4 ++++ ubicloud/SKILL.md | 2 +- ubicloud/sections/preamble.md | 7 +++---- write-skill/SKILL.md | 2 +- write-skill/sections/preamble.md | 7 +++---- 101 files changed, 199 insertions(+), 236 deletions(-) create mode 100644 changelog.d/gstack-reports-lead-with-failures.md create mode 100644 test/size-ledger/ship/gstack-reports-lead-with-failures.json diff --git a/autoplan/SKILL.md b/autoplan/SKILL.md index cb0cef8e0..1eda17e55 100644 --- a/autoplan/SKILL.md +++ b/autoplan/SKILL.md @@ -34,7 +34,7 @@ hooks: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/autoplan/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/autoplan/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/autoplan/sections/preamble.md b/autoplan/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/autoplan/sections/preamble.md +++ b/autoplan/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/canary/SKILL.md b/canary/SKILL.md index 643b31b0b..68d2b514f 100644 --- a/canary/SKILL.md +++ b/canary/SKILL.md @@ -19,7 +19,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/canary/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/canary/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/canary/sections/preamble.md b/canary/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/canary/sections/preamble.md +++ b/canary/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/captain/SKILL.md b/captain/SKILL.md index 6ab00a2b4..14d68cedb 100644 --- a/captain/SKILL.md +++ b/captain/SKILL.md @@ -26,7 +26,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/captain/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/captain/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/captain/sections/preamble.md b/captain/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/captain/sections/preamble.md +++ b/captain/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/captain/templates/READY.md b/captain/templates/READY.md index d31cda187..acbd4b84a 100644 --- a/captain/templates/READY.md +++ b/captain/templates/READY.md @@ -37,6 +37,8 @@ The lane's own lines, after the block: ```text pin: # the full head SHA, restated +Failed or skipped: +Also changed: screen choice: <--no-screen: cases ran focused and passed | --screen: > flaky: > review passes: (deferred: ) diff --git a/captain/templates/report.md b/captain/templates/report.md index bac7b35c8..39292b894 100644 --- a/captain/templates/report.md +++ b/captain/templates/report.md @@ -4,7 +4,7 @@ Written by the lane when it finishes. The controller reads this before chat. ## Outcome - + Unverified: diff --git a/capy/skill/gstack/factory.md b/capy/skill/gstack/factory.md index fa10da5d1..c6134b3a0 100644 --- a/capy/skill/gstack/factory.md +++ b/capy/skill/gstack/factory.md @@ -11,7 +11,7 @@ Your crew's CREW.md and recorded approvals decide who may merge, update a base, 7. **Measure before you act.** Run the real system on the merged tree you ship; a real run outranks a reviewed plan. Each number names its run, commit or uncapped query; say what you did not verify. 8. **Evals decide, and losses get reported.** Preregister bars, cap and stop rule before paying [/eval-plan]. Cheap proof (a keyless fake, a smoke) first. Controls: no-skill and placebo arms, clean decoys. Report use and pass apart; quote spend from the meter. A new part needs proof it helps or a dated test plan [ETHOS.md: 9. Earn Its Place]. 9. **Flakes: one rerun, a ledger, a daily fix.** Only the policy's rerun, never for a harness or infra red; a second red is real; never loosen a test. [crew/docs/captain.md: The daily flaky pass] -10. **Quiet.** Messages wake the Captain cold: send READY, a blocker, done or a Captain or owner call, else end_turn; no acks or progress notes. Broadcasts go in its notes. +10. **Quiet.** Messages wake the Captain cold: send READY, a blocker, done or a Captain or owner call, else end_turn; no acks or progress notes. Reports and replies open with `Failed or skipped:`, `Unverified:` and `Also changed:` lines (each item, or none), under 250 words; top 10 of long lists. Findings write-ups follow [ETHOS.md: 8. Explain the Number]. Broadcasts go in its notes. 11. **Ask once, plainly.** One ask per decision, explained like the owner is five: your Captain's name, then a lettered line each: "A: hit problem p, which happens when h. We could do 1) x, 2) y or 3) z. Recommendation x because why. A1 to take recommendation." "do it" takes every recommendation; "A2" takes option 2. Owner tasks: the exact link (in Capy: path and field), numbered steps, the reply word. Never re-ask a standing rule; check what you can yourself. [crew/docs/captain.md: The ask format] 12. **Fresh threads, closed out.** One thread per unit of work. A thread keeps its PR until it merges or closes and hands off only then, never mid-PR on a token count; past 1B own tokens it checks in with its Captain once and carries on. Standing threads that wake on every event (merge lanes, watchers) rotate at 500M tokens or 3 days [/runaway-stopper]. Crew threads never create threads: ask your Captain (Captain to Captain across projects), or report one you made by code for crew.add. Its Captain archives only its crew's threads, never one holding an open PR, successor or not (R91), or an open issue; a PR closes only as landed, superseded or dropped. Threads outside every crew are the owner's, even its automation's. Size reviews to the diff. [docs/ship-budget.md; crew/docs/captain.md: New threads join the crew] 13. **The owner's model preference.** `gstack-models resolve --role agent` picks agent models, except cross-family voices and eval subjects. [docs/model-policy.md: Your model preference] diff --git a/changelog.d/gstack-reports-lead-with-failures.md b/changelog.d/gstack-reports-lead-with-failures.md new file mode 100644 index 000000000..dd0dda6ec --- /dev/null +++ b/changelog.d/gstack-reports-lead-with-failures.md @@ -0,0 +1 @@ +- **gstack agents now open every status report with what failed, what they didn't verify and what else they changed.** The shared closing rule used to ask for "a few short lines". Now a report stays under 250 words and opens with three lines, `Failed or skipped:`, `Unverified:` and `Also changed:`, each naming every item or saying none. Lists longer than 10 get the top 10 and a count of the rest. Capy crews follow the same shape (factory.md point 10, the lane report and READY templates). On a 40-task dev split with Claude Opus 5.5, this wording raised the share of key facts disclosed in a report's first 250 words from 0.442 to 0.520 against the old closer, and in Capy lane reports from 0.552 to 0.604. Those are dev numbers, and the sealed held-out read is still to come (GST-1455). diff --git a/codex/SKILL.md b/codex/SKILL.md index 456134679..b19d6ef5a 100644 --- a/codex/SKILL.md +++ b/codex/SKILL.md @@ -20,7 +20,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/codex/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/codex/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/codex/sections/preamble.md b/codex/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/codex/sections/preamble.md +++ b/codex/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/context-restore/SKILL.md b/context-restore/SKILL.md index cafd34abd..68efd7af7 100644 --- a/context-restore/SKILL.md +++ b/context-restore/SKILL.md @@ -21,7 +21,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/context-restore/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/context-restore/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/context-restore/sections/preamble.md b/context-restore/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/context-restore/sections/preamble.md +++ b/context-restore/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/context-save/SKILL.md b/context-save/SKILL.md index 10a4e8ca6..28a4f9b64 100644 --- a/context-save/SKILL.md +++ b/context-save/SKILL.md @@ -21,7 +21,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/context-save/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/context-save/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/context-save/sections/preamble.md b/context-save/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/context-save/sections/preamble.md +++ b/context-save/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/design-consultation/SKILL.md b/design-consultation/SKILL.md index 7d545778d..adaa68b43 100644 --- a/design-consultation/SKILL.md +++ b/design-consultation/SKILL.md @@ -45,7 +45,7 @@ gbrain: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-consultation/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-consultation/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/design-consultation/sections/preamble.md b/design-consultation/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/design-consultation/sections/preamble.md +++ b/design-consultation/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/design-html/SKILL.md b/design-html/SKILL.md index 400cb285c..dee5afe1a 100644 --- a/design-html/SKILL.md +++ b/design-html/SKILL.md @@ -22,7 +22,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-html/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-html/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/design-html/sections/preamble.md b/design-html/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/design-html/sections/preamble.md +++ b/design-html/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/design-motion/SKILL.md b/design-motion/SKILL.md index f95627717..9d6601577 100644 --- a/design-motion/SKILL.md +++ b/design-motion/SKILL.md @@ -21,7 +21,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-motion/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-motion/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/design-motion/sections/preamble.md b/design-motion/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/design-motion/sections/preamble.md +++ b/design-motion/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/design-review/SKILL.md b/design-review/SKILL.md index 7713ee0f5..977491217 100644 --- a/design-review/SKILL.md +++ b/design-review/SKILL.md @@ -23,7 +23,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/design-review/sections/preamble.md b/design-review/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/design-review/sections/preamble.md +++ b/design-review/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/design-shotgun/SKILL.md b/design-shotgun/SKILL.md index cf888a102..a5fa30b71 100644 --- a/design-shotgun/SKILL.md +++ b/design-shotgun/SKILL.md @@ -39,7 +39,7 @@ gbrain: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-shotgun/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/design-shotgun/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/design-shotgun/sections/preamble.md b/design-shotgun/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/design-shotgun/sections/preamble.md +++ b/design-shotgun/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/devex-review/SKILL.md b/devex-review/SKILL.md index d353ba1bb..d2b158dc8 100644 --- a/devex-review/SKILL.md +++ b/devex-review/SKILL.md @@ -21,7 +21,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/devex-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/devex-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/devex-review/sections/preamble.md b/devex-review/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/devex-review/sections/preamble.md +++ b/devex-review/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/document-generate/SKILL.md b/document-generate/SKILL.md index a1c9da8ba..c8d9cdbcb 100644 --- a/document-generate/SKILL.md +++ b/document-generate/SKILL.md @@ -25,7 +25,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/document-generate/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/document-generate/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/document-generate/sections/preamble.md b/document-generate/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/document-generate/sections/preamble.md +++ b/document-generate/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/document-release/SKILL.md b/document-release/SKILL.md index 3d1a91dc1..8047fc11e 100644 --- a/document-release/SKILL.md +++ b/document-release/SKILL.md @@ -21,7 +21,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/document-release/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/document-release/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/document-release/sections/preamble.md b/document-release/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/document-release/sections/preamble.md +++ b/document-release/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/eval-plan/SKILL.md b/eval-plan/SKILL.md index c7665440e..e5a34067e 100644 --- a/eval-plan/SKILL.md +++ b/eval-plan/SKILL.md @@ -24,7 +24,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/eval-plan/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/eval-plan/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/eval-plan/sections/preamble.md b/eval-plan/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/eval-plan/sections/preamble.md +++ b/eval-plan/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/health/SKILL.md b/health/SKILL.md index 0df11a75b..2a9e6b7f7 100644 --- a/health/SKILL.md +++ b/health/SKILL.md @@ -21,7 +21,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/health/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/health/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/health/sections/preamble.md b/health/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/health/sections/preamble.md +++ b/health/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/ios-clean/SKILL.md b/ios-clean/SKILL.md index 80b44691d..d76f480b1 100644 --- a/ios-clean/SKILL.md +++ b/ios-clean/SKILL.md @@ -20,7 +20,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-clean/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-clean/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/ios-clean/sections/preamble.md b/ios-clean/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/ios-clean/sections/preamble.md +++ b/ios-clean/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/ios-design-review/SKILL.md b/ios-design-review/SKILL.md index 3d630daa3..1d47b7219 100644 --- a/ios-design-review/SKILL.md +++ b/ios-design-review/SKILL.md @@ -19,7 +19,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-design-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-design-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/ios-design-review/sections/preamble.md b/ios-design-review/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/ios-design-review/sections/preamble.md +++ b/ios-design-review/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/ios-fix/SKILL.md b/ios-fix/SKILL.md index af59d3244..8771cf89c 100644 --- a/ios-fix/SKILL.md +++ b/ios-fix/SKILL.md @@ -21,7 +21,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-fix/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-fix/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/ios-fix/sections/preamble.md b/ios-fix/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/ios-fix/sections/preamble.md +++ b/ios-fix/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/ios-qa/SKILL.md b/ios-qa/SKILL.md index f34efd275..712a36166 100644 --- a/ios-qa/SKILL.md +++ b/ios-qa/SKILL.md @@ -24,7 +24,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-qa/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-qa/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/ios-qa/sections/preamble.md b/ios-qa/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/ios-qa/sections/preamble.md +++ b/ios-qa/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/ios-sync/SKILL.md b/ios-sync/SKILL.md index 1c49407a5..8a75aebfc 100644 --- a/ios-sync/SKILL.md +++ b/ios-sync/SKILL.md @@ -21,7 +21,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-sync/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ios-sync/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/ios-sync/sections/preamble.md b/ios-sync/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/ios-sync/sections/preamble.md +++ b/ios-sync/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/land-and-deploy/SKILL.md b/land-and-deploy/SKILL.md index 08a7ced13..7a12e351b 100644 --- a/land-and-deploy/SKILL.md +++ b/land-and-deploy/SKILL.md @@ -20,7 +20,7 @@ disable-model-invocation: true ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/land-and-deploy/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/land-and-deploy/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/land-and-deploy/sections/preamble.md b/land-and-deploy/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/land-and-deploy/sections/preamble.md +++ b/land-and-deploy/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/landing-report/SKILL.md b/landing-report/SKILL.md index b07a98cf3..616a9eb4c 100644 --- a/landing-report/SKILL.md +++ b/landing-report/SKILL.md @@ -20,7 +20,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/landing-report/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/landing-report/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/landing-report/sections/preamble.md b/landing-report/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/landing-report/sections/preamble.md +++ b/landing-report/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/learn/SKILL.md b/learn/SKILL.md index 9c27e322b..52b620d8a 100644 --- a/learn/SKILL.md +++ b/learn/SKILL.md @@ -21,7 +21,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/learn/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/learn/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/learn/sections/preamble.md b/learn/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/learn/sections/preamble.md +++ b/learn/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/modernize/SKILL.md b/modernize/SKILL.md index ed48ce690..e73b7b965 100644 --- a/modernize/SKILL.md +++ b/modernize/SKILL.md @@ -23,7 +23,7 @@ disable-model-invocation: true ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/modernize/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/modernize/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/modernize/sections/preamble.md b/modernize/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/modernize/sections/preamble.md +++ b/modernize/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/office-hours/SKILL.md b/office-hours/SKILL.md index c40b91938..374b09e8a 100644 --- a/office-hours/SKILL.md +++ b/office-hours/SKILL.md @@ -49,7 +49,7 @@ gbrain: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/office-hours/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/office-hours/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/office-hours/sections/preamble.md b/office-hours/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/office-hours/sections/preamble.md +++ b/office-hours/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/pair-agent/SKILL.md b/pair-agent/SKILL.md index 457205758..cb3d8c48e 100644 --- a/pair-agent/SKILL.md +++ b/pair-agent/SKILL.md @@ -18,7 +18,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/pair-agent/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/pair-agent/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/pair-agent/sections/preamble.md b/pair-agent/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/pair-agent/sections/preamble.md +++ b/pair-agent/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/plan-ceo-review/SKILL.md b/plan-ceo-review/SKILL.md index 69a4dec1e..f34a1f223 100644 --- a/plan-ceo-review/SKILL.md +++ b/plan-ceo-review/SKILL.md @@ -51,7 +51,7 @@ gbrain: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-ceo-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-ceo-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/plan-ceo-review/sections/preamble.md b/plan-ceo-review/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/plan-ceo-review/sections/preamble.md +++ b/plan-ceo-review/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/plan-design-review/SKILL.md b/plan-design-review/SKILL.md index 61b7a6363..fe6c303e8 100644 --- a/plan-design-review/SKILL.md +++ b/plan-design-review/SKILL.md @@ -51,7 +51,7 @@ Recommendation: A when a branch diff exists, otherwise B. Reply with A, B, or C. ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-design-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-design-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (after scope gate) diff --git a/plan-design-review/sections/preamble.md b/plan-design-review/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/plan-design-review/sections/preamble.md +++ b/plan-design-review/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/plan-devex-review/SKILL.md b/plan-devex-review/SKILL.md index cbe3a2fe1..d815672a5 100644 --- a/plan-devex-review/SKILL.md +++ b/plan-devex-review/SKILL.md @@ -21,7 +21,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-devex-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-devex-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/plan-devex-review/sections/preamble.md b/plan-devex-review/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/plan-devex-review/sections/preamble.md +++ b/plan-devex-review/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/plan-eng-review/SKILL.md b/plan-eng-review/SKILL.md index 745640604..79714f317 100644 --- a/plan-eng-review/SKILL.md +++ b/plan-eng-review/SKILL.md @@ -63,7 +63,7 @@ Keep the reviewed target fixed when selecting the report destination. ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-eng-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-eng-review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (after scope gate) diff --git a/plan-eng-review/sections/preamble.md b/plan-eng-review/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/plan-eng-review/sections/preamble.md +++ b/plan-eng-review/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/plan-tune/SKILL.md b/plan-tune/SKILL.md index d30cacddc..d0a412971 100644 --- a/plan-tune/SKILL.md +++ b/plan-tune/SKILL.md @@ -25,7 +25,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-tune/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/plan-tune/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/plan-tune/sections/preamble.md b/plan-tune/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/plan-tune/sections/preamble.md +++ b/plan-tune/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/qa-only/SKILL.md b/qa-only/SKILL.md index 4854a744a..23f911322 100644 --- a/qa-only/SKILL.md +++ b/qa-only/SKILL.md @@ -19,7 +19,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/qa-only/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/qa-only/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/qa-only/sections/preamble.md b/qa-only/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/qa-only/sections/preamble.md +++ b/qa-only/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/qa/SKILL.md b/qa/SKILL.md index 38d95e46e..c1fb54fed 100644 --- a/qa/SKILL.md +++ b/qa/SKILL.md @@ -22,7 +22,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/qa/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/qa/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/qa/sections/preamble.md b/qa/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/qa/sections/preamble.md +++ b/qa/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/retro/SKILL.md b/retro/SKILL.md index 87f44852c..c4f0c6240 100644 --- a/retro/SKILL.md +++ b/retro/SKILL.md @@ -39,7 +39,7 @@ gbrain: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/retro/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/retro/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/retro/sections/preamble.md b/retro/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/retro/sections/preamble.md +++ b/retro/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/review/SKILL.md b/review/SKILL.md index a1809271b..eb7b4cabe 100644 --- a/review/SKILL.md +++ b/review/SKILL.md @@ -50,7 +50,7 @@ Finish only when each of these exists (contract v1): ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/review/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/review/sections/preamble.md b/review/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/review/sections/preamble.md +++ b/review/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/scripts/resolvers/preamble/generate-voice-directive.ts b/scripts/resolvers/preamble/generate-voice-directive.ts index 07b3e0279..ccb72a3ae 100644 --- a/scripts/resolvers/preamble/generate-voice-directive.ts +++ b/scripts/resolvers/preamble/generate-voice-directive.ts @@ -31,8 +31,7 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: \`Failed or skipped:\` (anything that failed, even if fixed, was skipped or never ran), \`Unverified:\` (claims no output confirmed), \`Also changed:\` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned.`; +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none."`; } diff --git a/setup-deploy/SKILL.md b/setup-deploy/SKILL.md index c06932f5b..184cf047c 100644 --- a/setup-deploy/SKILL.md +++ b/setup-deploy/SKILL.md @@ -21,7 +21,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/setup-deploy/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/setup-deploy/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/setup-deploy/sections/preamble.md b/setup-deploy/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/setup-deploy/sections/preamble.md +++ b/setup-deploy/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/setup-gbrain/SKILL.md b/setup-gbrain/SKILL.md index 7b4355c2d..1d3c9563d 100644 --- a/setup-gbrain/SKILL.md +++ b/setup-gbrain/SKILL.md @@ -23,7 +23,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/setup-gbrain/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/setup-gbrain/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/setup-gbrain/sections/preamble.md b/setup-gbrain/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/setup-gbrain/sections/preamble.md +++ b/setup-gbrain/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/ship/SKILL.md b/ship/SKILL.md index d64101278..6f0453a16 100644 --- a/ship/SKILL.md +++ b/ship/SKILL.md @@ -24,7 +24,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ship/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ship/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/ship/sections/preamble.md b/ship/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/ship/sections/preamble.md +++ b/ship/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/skillify/SKILL.md b/skillify/SKILL.md index 10477485e..cb169962f 100644 --- a/skillify/SKILL.md +++ b/skillify/SKILL.md @@ -19,7 +19,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/skillify/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/skillify/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/skillify/sections/preamble.md b/skillify/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/skillify/sections/preamble.md +++ b/skillify/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/spec/SKILL.md b/spec/SKILL.md index 1529f557a..016aa3620 100644 --- a/spec/SKILL.md +++ b/spec/SKILL.md @@ -22,7 +22,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/spec/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/spec/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/spec/sections/preamble.md b/spec/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/spec/sections/preamble.md +++ b/spec/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/sync-gbrain/SKILL.md b/sync-gbrain/SKILL.md index 354f96e6f..63e4054b7 100644 --- a/sync-gbrain/SKILL.md +++ b/sync-gbrain/SKILL.md @@ -22,7 +22,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/sync-gbrain/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/sync-gbrain/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/sync-gbrain/sections/preamble.md b/sync-gbrain/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/sync-gbrain/sections/preamble.md +++ b/sync-gbrain/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/test-audit/SKILL.md b/test-audit/SKILL.md index 090bbb322..48f3f4578 100644 --- a/test-audit/SKILL.md +++ b/test-audit/SKILL.md @@ -21,7 +21,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/test-audit/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/test-audit/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/test-audit/sections/preamble.md b/test-audit/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/test-audit/sections/preamble.md +++ b/test-audit/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/test/capy-skill-wrapper.test.ts b/test/capy-skill-wrapper.test.ts index 9ae7afb50..2f4e31031 100644 --- a/test/capy-skill-wrapper.test.ts +++ b/test/capy-skill-wrapper.test.ts @@ -680,14 +680,16 @@ describe('capy skill: factory.md, the auto-loaded "how we work" page', () => { return [...new Set(problems)]; } - test('it fits the 4,200-byte budget, and every citation, heading and skill it names resolves', () => { + test('it fits the 4,400-byte budget, and every citation, heading and skill it names resolves', () => { const text = fs.readFileSync(FACTORY, 'utf8'); // 3,300 since #3634; raised to 3,450 (the page's new size, rounded up to the next 25 bytes) for point 12's // R68 clause (Garry, 2026-11-03: crew threads ask their Captain for new threads). Raise it only by what a rule needs. // 3,475 for the crew/ prefix on its four Captain-manual citations (GST-1833 moved docs/captain.md to crew/docs/; no rule text changed). // 3,650 for point 2's one-open-PR-per-lane rule and point 12's no-orphan and close-reason rule (GST-2077, Garry's "do it" 2026-11-09: lessons from gbrain's merge queue). // 4,200 for point 12's keep-the-PR-to-the-end and standing-thread rotation clause and point 14's cap of 10 finished PRs before a new fix thread (Garry's "do it" 2026-11-09, relayed by FOUN-32). - expect(Buffer.byteLength(text)).toBeLessThanOrEqual(4200); + // 4,400 for point 10's report shape, Failed or skipped / Unverified / Also changed lines first, and its pointer to ETHOS.md section 8 for + // findings write-ups (GST-1455, Garry's "do it" 2026-11-05 on R73; the clause is the wording GEVA-2's dev climb measured, so it is not trimmed). + expect(Buffer.byteLength(text)).toBeLessThanOrEqual(4400); expect(text).toMatch(/^# How we work/); expect(citationProblems(text)).toEqual([]); }); diff --git a/test/fixtures/golden/agy-ship-SKILL.md b/test/fixtures/golden/agy-ship-SKILL.md index b84aa8045..d38fbc2b6 100644 --- a/test/fixtures/golden/agy-ship-SKILL.md +++ b/test/fixtures/golden/agy-ship-SKILL.md @@ -12,7 +12,7 @@ description: | ## Shared preamble (load first) -This skill follows gstack shared preamble `dc6688a6791f` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `sections/preamble.md` relative to the installed `gstack-ship` SKILL.md directory in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `673baf289542` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `sections/preamble.md` relative to the installed `gstack-ship` SKILL.md directory in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/test/fixtures/golden/claude-ship-SKILL.md b/test/fixtures/golden/claude-ship-SKILL.md index d64101278..6f0453a16 100644 --- a/test/fixtures/golden/claude-ship-SKILL.md +++ b/test/fixtures/golden/claude-ship-SKILL.md @@ -24,7 +24,7 @@ triggers: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ship/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ship/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/test/fixtures/golden/codex-ship-SKILL.md b/test/fixtures/golden/codex-ship-SKILL.md index d535a44b4..002db2ba4 100644 --- a/test/fixtures/golden/codex-ship-SKILL.md +++ b/test/fixtures/golden/codex-ship-SKILL.md @@ -12,7 +12,7 @@ description: | ## Shared preamble (load first) -This skill follows gstack shared preamble `1c66f530e65a` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `sections/preamble.md` relative to the installed `gstack-ship` SKILL.md directory in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `0876a471d87b` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `sections/preamble.md` relative to the installed `gstack-ship` SKILL.md directory in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/test/fixtures/golden/factory-ship-SKILL.md b/test/fixtures/golden/factory-ship-SKILL.md index 31f0a060c..1a11689c3 100644 --- a/test/fixtures/golden/factory-ship-SKILL.md +++ b/test/fixtures/golden/factory-ship-SKILL.md @@ -14,7 +14,7 @@ disable-model-invocation: true ## Shared preamble (load first) -This skill follows gstack shared preamble `fb2ce44dd505` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `sections/preamble.md` relative to the installed `gstack-ship` SKILL.md directory in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `b517469fdbdb` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `sections/preamble.md` relative to the installed `gstack-ship` SKILL.md directory in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/test/size-ledger/ship/gstack-reports-lead-with-failures.json b/test/size-ledger/ship/gstack-reports-lead-with-failures.json new file mode 100644 index 000000000..8f317672f --- /dev/null +++ b/test/size-ledger/ship/gstack-reports-lead-with-failures.json @@ -0,0 +1,4 @@ +{ + "bytes": -4, + "why": "GST-1455: the shared closer opens status reports with failed-or-skipped, unverified and also-changed lines (-4 B)" +} diff --git a/ubicloud/SKILL.md b/ubicloud/SKILL.md index 6274dbed6..c86b42b5c 100644 --- a/ubicloud/SKILL.md +++ b/ubicloud/SKILL.md @@ -19,7 +19,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ubicloud/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/ubicloud/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/ubicloud/sections/preamble.md b/ubicloud/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/ubicloud/sections/preamble.md +++ b/ubicloud/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery diff --git a/write-skill/SKILL.md b/write-skill/SKILL.md index 85fc30a8b..7d4986e2a 100644 --- a/write-skill/SKILL.md +++ b/write-skill/SKILL.md @@ -22,7 +22,7 @@ allowed-tools: ## Shared preamble (load first) -This skill follows gstack shared preamble `227aca067dd3` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/write-skill/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. +This skill follows gstack shared preamble `d56c99c92144` (AskUserQuestion format, voice, writing, completeness and completion rules). If your Read of that file, headed by this id, is still in context, do not read it again; otherwise Read `~/.claude/skills/gstack/write-skill/sections/preamble.md` in full before the command below. A section or subsection here with a shared heading replaces the shared one. ## Preamble (run first) diff --git a/write-skill/sections/preamble.md b/write-skill/sections/preamble.md index 9efe49a9d..a7fb731da 100644 --- a/write-skill/sections/preamble.md +++ b/write-skill/sections/preamble.md @@ -1,6 +1,6 @@ -# gstack shared preamble 227aca067dd3 +# gstack shared preamble d56c99c92144 ## Plan Mode Safe Operations @@ -171,10 +171,9 @@ GStack voice: Garry-shaped product and engineering judgment. Good: "auth.ts:47 returns undefined when the session cookie expires. Users hit a white screen. Fix: add a null check and redirect to /login. Two lines." Bad: "I've identified a potential issue in the authentication flow that may cause problems under certain conditions." -**Bounded closer.** After completing work, report in at most a few short lines: what changed, what was skipped, what to watch. No feature tours or unrequested design notes. Exempt: decision briefs, completion-status blocks, requested explanations, and a skill's mandated report (/qa-only, /plan-*-review, /retro, /document-generate). The rule limits prose around the deliverable, never the deliverable. +**Bounded closer.** A status report (after work or when asked) stays under 250 words and opens with three lines, each item named, or none: `Failed or skipped:` (anything that failed, even if fixed, was skipped or never ran), `Unverified:` (claims no output confirmed), `Also changed:` (config, CI, files outside the ask). Then what changed and ran; claim nothing more. Over 10 items: top 10, the rest counted, an offer to list. Exempt: decision briefs, requested explanations, mandated reports (/qa-only, /plan-*-review, /retro, /document-generate). -Good closer: "Renamed the flag in 3 files, regenerated docs, tests green. Skipped the CLI alias (unused since v1.2); watch the Windows job." -Bad closer: a tour of every edit, a restatement of the plan, and three paragraphs justifying choices nobody questioned. +Good: "Failed or skipped: lint (fixed), 2 e2e tests (no key). Unverified: Windows alias. Also changed: none." ## Context Recovery