CalvinBackup/gstack
1
0
Fork 0
mirror of https://github.com/garrytan/gstack.git synced 2026-05-02 03:35:09 +02:00
Code Issues Packages Projects Releases Wiki Activity
Files
main
gstack/AGENTS.md
T
Garry Tan 6e1625c0d7 v1.25.0.0 fix: AskUserQuestion resolves to host MCP variant when native is disallowed (#1287) 
* test(harness): plumb extraArgs and auto_decided outcome through PTY runner

runPlanSkillObservation now accepts extraArgs that pass through to
launchClaudePty (which already supported them at the lower level), and
exposes a new 'auto_decided' outcome detected via isAutoDecidedVisible
when the AUTO_DECIDE preamble template fires (Auto-decided ... (your
preference)).

Both pieces are needed for the v1.21+ AskUserQuestion-blocked regression
tests in the next commit. Detection order is deliberate: 'asked' (rendered
numbered list) wins over 'auto_decided' (text only, no list), which wins
over 'plan_ready' so the auto-decide evidence isn't masked by a downstream
plan-mode confirmation.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* test(e2e): add AskUserQuestion-blocked regression cases for 6 plan-mode skills

Conductor launches Claude Code with --disallowedTools AskUserQuestion
--permission-mode default --permission-prompt-tool stdio (verified by
inspecting the live conductor claude process via ps -p ... -o args=).
Native AskUserQuestion is removed from the model's tool registry; without
fallback guidance the plan-mode skills (plan-ceo-review, plan-eng-review,
plan-design-review, plan-devex-review, autoplan, office-hours) silently
proceed and never surface decisions to the user.

Adds 6 gate-tier real-PTY regression cases:

  - 4 inline test cases inside the existing plan-X-review-plan-mode.test
    files, each exercising the same skill with extraArgs ['--disallowedTools',
    'AskUserQuestion'] and asserting outcome === 'asked'. plan-design-review
    keeps the ['asked', 'plan_ready'] envelope (legitimate short-circuit on
    no-UI-scope) but explicitly fails on 'auto_decided'.
  - 2 standalone test files for autoplan + office-hours (which had no prior
    plan-mode test). autoplan asserts the FIRST non-auto-decided gate fires
    (Phase 1 premise confirmation) — autoplan auto-decides intermediate
    questions BY DESIGN.

Touchfile entries:
  - autoplan-auto-mode + office-hours-auto-mode added to E2E_TOUCHFILES +
    E2E_TIERS (gate)
  - existing plan-X-review-plan-mode entries gain question-tuning.ts and
    generate-ask-user-format.ts touchfile deps so AUTO_DECIDE-related
    resolver changes correctly invalidate the regression tests
  - touchfiles.test.ts count updated 18 -> 19 to cover the autoplan
    touchfile dependency on plan-ceo-review/**

Filenames retain `auto-mode` for branch-history continuity. Auto-mode (the
AUTO_DECIDE preamble path when QUESTION_TUNING=true) is a related but
distinct silencing mechanism; both share the same fix surface in the
preamble.

These tests are expected to FAIL on this branch until the fix lands. The
failure is the receipt for the regression.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(preamble): teach the model to prefer mcp__*__AskUserQuestion when registered

When a host launches Claude Code with --disallowedTools AskUserQuestion
(Conductor does this by default — verified via ps on the live conductor
claude process), the native AskUserQuestion tool is removed from the
model's tool registry. Skill templates that say "call AskUserQuestion"
silently fail in that environment: the model can't ask, the user never
sees the question, the skill auto-proceeds without input.

The fix is preamble guidance, not a skill-template change:

  generate-ask-user-format.ts: new "Tool resolution" section at the top
  of the AskUserQuestion Format block. Tells the model that
  "AskUserQuestion" can resolve to two tools at runtime — the host MCP
  variant (e.g. mcp__conductor__AskUserQuestion, registered when the
  host injects it) and the native tool — and to PREFER any
  mcp__*__AskUserQuestion variant. Same questions/options shape; same
  decision-brief format. If neither variant is callable, fall back to
  writing a "## Decisions to confirm" section into the plan file plus
  ExitPlanMode (the native plan-mode confirmation surfaces it). Never
  silently auto-decide.

  generate-completion-status.ts: the plan-mode-info block (preamble
  position 1) now explicitly notes that AskUserQuestion satisfies plan
  mode's end-of-turn requirement for "any variant" and points at the
  Tool resolution section for the fallback path.

This puts the resolution rule in front of every tier-≥2 skill via the
preamble, so plan-mode review skills (plan-ceo-review, plan-eng-review,
plan-design-review, plan-devex-review, autoplan, office-hours) all gain
the fix without per-template surgery.

Includes regenerated SKILL.md files for all 41 skills + the 3 host-ship
golden fixtures used by test/host-config.test.ts.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* test(periodic): AUTO_DECIDE opt-in preserved under Conductor flags

Periodic-tier eval that exercises the legitimate /plan-tune AUTO_DECIDE
path under the same flags Conductor uses (--disallowedTools
AskUserQuestion). Confirms the new Tool resolution preamble doesn't trip
opt-in users: when the user has set a never-ask preference for a
question, the model should auto-pick (outcome 'auto_decided' or
'plan_ready') rather than surface the prompt.

Setup runs in an isolated GSTACK_HOME tmpdir — never touches the user's
real ~/.gstack state. Writes question_tuning=true + a never-ask
preference for plan-ceo-review-mode (source: 'plan-tune', which bypasses
the inline-user origin gate). Spawns claude with
--disallowedTools AskUserQuestion in plan mode, runs /plan-ceo-review,
asserts outcome is NOT 'asked' (i.e., the model honored the preference).

Periodic tier because AUTO_DECIDE behavior depends on the model adhering
to the QUESTION_TUNING preamble injection — non-deterministic, weekly
cron is the right cadence rather than CI gating.

Touchfiles cover the AUTO_DECIDE-bearing resolvers + the question-tuning
binaries the test setup invokes. touchfiles.test.ts count updates 19 ->
20 because auto-decide-preserved also depends on plan-ceo-review/**.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* v1.21.0.0: AskUserQuestion resolves to host MCP variant when native is disallowed

MINOR scale per scale-aware bumps in CLAUDE.md: substantial coordinated
multi-file change (preamble fix + new test infrastructure + 6 gate-tier
regression cases + 1 periodic eval) and a user-visible regression fix
that affects every plan-mode review skill running under Conductor's
default flag set.

User originally targeted v1.21.2.0; landing as v1.21.0.0 since this is
the first 1.21.x release on main and there's no prior 1.21.0.0/1.21.1.0
to skip past. Adjust at /ship time if a different number is preferred.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* test(harness): fix detection order + whitespace-tolerant pattern matching

Two bugs surfaced when validating the v1.21 fix end-to-end:

1. PlanSkillObservation outcome detection ran 'asked' (any numbered
   options list) BEFORE 'plan_ready'. Plan-mode's "Ready to execute?"
   confirmation IS a numbered options list (1=auto, 2=manual, ...), so
   any skill that successfully reached the native confirmation got
   misclassified as 'asked'. Reorder: 'auto_decided' (most specific,
   requires AUTO_DECIDE annotation) > 'plan_ready' (next, requires the
   "ready to execute" stem) > 'asked' (any remaining numbered list).

2. isPlanReadyVisible and isAutoDecidedVisible regexes only matched
   spaced forms ("ready to execute", "(your preference)"). stripAnsi
   removes cursor-positioning escapes (`\x1b[40C`) entirely instead of
   replacing them with spaces, so the same text can render as
   "readytoexecute" or "(yourpreference)". Both detectors now test the
   spaced form first, fall through to a whitespace-collapsed comparison.
   Inline unit smoke confirms both forms match.

Updates to the 5 strict 'asked' regression test cases (plan-ceo,
plan-eng, plan-devex, autoplan, office-hours): with the detection order
corrected, the model's plan-file fallback flow legitimately lands at
'plan_ready' instead of 'asked'. Pass envelope expanded to ['asked',
'plan_ready'] (matching plan-design-review's existing pattern). Failure
signals tightened to include 'auto_decided' (catches AUTO_DECIDE without
opt-in) plus the standard silent_write/exited/timeout. plan-design was
already on this contract from v1.21's first commit, no change needed.

The expanded envelope is correct: under --disallowedTools AskUserQuestion
the Tool resolution preamble routes the question through plan-mode's
native "Ready to execute?" surface — the user still sees the decision,
just via the plan-file flow rather than a numbered prompt.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* test(harness): require ## Decisions section under --disallowedTools plan_ready

Adversarial review (during /ship Step 11) found that the previous gate-test
envelope ['asked', 'plan_ready'] for the AskUserQuestion-blocked regression
cases accepted the bug they exist to catch: a model that silently skips
Step 0 entirely (writes a plan with no questions, no `## Decisions to
confirm` section, just ExitPlanModes) reaches plan_ready and passes.

The fix tightens the contract in two layers:

1. Harness: PlanSkillObservation gains a `planFile?: string` field
   populated when outcome is plan_ready. extractPlanFilePath() walks the
   visible TTY buffer for "Plan saved to:", "Plan file:", or
   ".claude/plans/<name>.md" patterns and resolves tilde to absolute.
   planFileHasDecisionsSection() reads the resolved file and returns true
   if it contains a `## Decisions` heading (any form: "to confirm",
   "needed", etc.).

2. Tests: 5 of 6 regression cases now require, when outcome is plan_ready,
   that obs.planFile is set AND planFileHasDecisionsSection returns true.
   Otherwise the test fails with a "Step 0 was silently skipped" diagnosis.
   plan-design-review remains the sole exception — it legitimately
   short-circuits to plan_ready on no-UI-scope branches and we have no
   deterministic way to distinguish that from a silent skip.

This closes the loophole the adversarial review identified. The fix
preamble flow already tells the model to write `## Decisions to confirm`
when neither AUQ variant is callable — now the test verifies the model
actually did it.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix(harness): anchor extractPlanFilePath path captures on /Users|~|/home|/var|/tmp

Adversarial-tightened gate sweep surfaced a real bug in the path
extraction: stripAnsi collapses whitespace via cursor-positioning escape
removal, so "yet at /Users/..." in the visible buffer becomes
"yetat/Users/..." with no space between. The previous fallback pattern
`(~?\/?\S*\.claude\/plans\/[\w-]+\.md)` greedily matched non-whitespace
characters BEFORE the path, producing `yetat/Users/garrytan/.claude/...`
which then fails fs.readFileSync.

Fix: every regex now requires the path to START at a known path-anchor:
`~/`, `/Users/`, `/home/`, `/var/`, `/tmp/`, or `./`. Earlier
non-whitespace runs can't be glommed in.

Verified against the failing fixture (`yetat/Users/...`) plus the four
canonical render forms ("Plan saved to:", "Plan file:", `·`-decorated
ctrl-g hint, and the bare fallback).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-01 08:45:36 -07:00

5.8 KiB
Raw Permalink Blame History

gstack — AI Engineering Workflow

gstack is a collection of SKILL.md files that give AI agents structured roles for software development. Each skill is a specialist: CEO reviewer, eng manager, designer, QA lead, release engineer, debugger, and more.

Available skills

Skills live in .agents/skills/ (or ~/.claude/skills/gstack/ on Claude Code). Invoke them by name (e.g., /office-hours).

Plan-mode reviews

Skill What it does
/office-hours Start here. Reframes your product idea before you write code.
/plan-ceo-review CEO-level review: find the 10-star product in the request.
/plan-eng-review Lock architecture, data flow, edge cases, and tests.
/plan-design-review Rate each design dimension 0-10, explain what a 10 looks like.
/plan-devex-review DX-mode review: TTHW, magical moments, friction points, persona traces.
/plan-tune Self-tune AskUserQuestion sensitivity per question.
/autoplan One command runs CEO → design → eng → DX review.
/design-consultation Build a complete design system from scratch.

Implementation + review

Skill What it does
/review Pre-landing PR review. Finds bugs that pass CI but break in prod.
/codex Second opinion via OpenAI Codex. Review, challenge, or consult modes.
/investigate Systematic root-cause debugging. No fixes without investigation.
/design-review Live-site visual audit + fix loop with atomic commits.
/design-shotgun Generate multiple AI design variants, comparison board, iterate.
/design-html Generate production-quality Pretext-native HTML/CSS.
/devex-review Live developer experience audit (TTHW measured against the real flow).
/qa Open a real browser, find bugs, fix them, re-verify.
/qa-only Same methodology as /qa but report only — no code changes.
/scrape Pull data from a web page. First call prototypes; codified call runs in ~200ms.
/skillify Codify the most recent successful /scrape flow into a permanent browser-skill.

Release + deploy

Skill What it does
/ship Run tests, review, push, open PR. Workspace-aware version queue.
/land-and-deploy Merge the PR, wait for CI and deploy, verify production health.
/canary Post-deploy monitoring loop using the browse daemon.
/landing-report Read-only dashboard for the workspace-aware ship queue.
/document-release Update all docs to match what you just shipped.
/setup-deploy One-time deploy config detection (Fly.io, Render, Vercel, etc.).
/gstack-upgrade Update gstack to the latest version.

Operational + memory

Skill What it does
/context-save Save working context (git state, decisions, remaining work).
/context-restore Resume from a saved context, even across Conductor workspaces.
/learn Manage what gstack learned across sessions.
/retro Weekly retro with per-person breakdowns and shipping streaks.
/health Code quality dashboard (type checker, linter, tests, dead code).
/benchmark Performance regression detection (page load, Core Web Vitals).
/benchmark-models Cross-model benchmark for skills (Claude, GPT, Gemini side-by-side).
/cso OWASP Top 10 + STRIDE security audit.
/setup-gbrain Set up gbrain for cross-machine session memory sync.

Browser + agent integration

Skill What it does
/browse Headless browser — real Chromium, real clicks, ~100ms/command.
/open-gstack-browser Launch the visible GStack Browser with sidebar + stealth.
/setup-browser-cookies Import cookies from your real browser for authenticated testing.
/pair-agent Pair a remote AI agent (OpenClaw, Codex, etc.) with your browser.

Safety + scoping

Skill What it does
/careful Warn before destructive commands (rm -rf, DROP TABLE, force-push).
/freeze Lock edits to one directory. Hard block, not just a warning.
/guard Activate both careful + freeze at once.
/unfreeze Remove directory edit restrictions.
/make-pdf Turn any markdown file into a publication-quality PDF.

Build commands

bun install              # install dependencies
bun test                 # run free tests (no API spend)
bun run test:windows     # curated Windows-safe subset (runs on windows-latest)
bun run build            # generate docs + compile binaries
bun run gen:skill-docs   # regenerate SKILL.md files from templates
bun run skill:check      # health dashboard for all skills

Platform support

  • macOS + Linux: full test suite supported.
  • Windows: curated Windows-safe subset runs on windows-latest via the windows-free-tests CI job. Setup script (./setup) requires Git Bash or MSYS today; native PowerShell support is a future expansion. The bin/gstack-paths helper resolves state roots through CLAUDE_PLUGIN_DATA / GSTACK_HOME so plugin installs work on every platform.

Key conventions

  • SKILL.md files are generated from .tmpl templates. Edit the template, not the output.
  • Run bun run gen:skill-docs --host codex to regenerate Codex-specific output.
  • The browse binary provides headless browser access. Use $B <command> in skills.
  • Safety skills (careful, freeze, guard) use inline advisory prose — always confirm before destructive operations.
  • State paths resolve via bin/gstack-paths (sourced via eval "$(...)"). Honors GSTACK_HOME, CLAUDE_PLUGIN_DATA, CLAUDE_PLANS_DIR.
  • The claude CLI binary resolves via browse/src/claude-bin.ts (Bun.which() + GSTACK_CLAUDE_BIN override). Set GSTACK_CLAUDE_BIN=wsl plus GSTACK_CLAUDE_BIN_ARGS='["claude"]' to run Claude through WSL on Windows.
Reference in New Issue View Git Blame Copy Permalink