From 7384ca0d485ba417dc8fa14692cdd7afb254b9c3 Mon Sep 17 00:00:00 2001 From: garrytan Date: Tue, 3 Nov 2026 02:38:30 +0000 Subject: [PATCH] fix(skills): Done-means list leaves out steps a host suppresses and allows the small-diff skip (review fix round) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> --- review/SKILL.md | 2 +- scripts/resolvers/contract-done-means.ts | 15 ++++++++++++--- test/fixtures/skill-md-size-allowlist.json | 18 +++++++++--------- test/helpers/carve-guards.ts | 4 ++-- test/skill-descriptions.test.ts | 14 ++++++++++++-- 5 files changed, 36 insertions(+), 17 deletions(-) diff --git a/review/SKILL.md b/review/SKILL.md index e15a15b19..66f4ba8be 100644 --- a/review/SKILL.md +++ b/review/SKILL.md @@ -42,7 +42,7 @@ contract: Finish only when each of these exists (contract v1): -- `specialists-dispatched`: the specialist selection printed (`Dispatching N specialists:` ...), then one Agent dispatch per specialist it names. +- `specialists-dispatched`: the specialist selection printed (`Dispatching N specialists:` ...), then one Agent dispatch per specialist it names (none when a small diff skips them). - `adversarial-pass`: an Agent dispatch for the adversarial pass, its prompt including "Think like an attacker and a chaos engineer". - `review-report`: a `Pre-Landing Review` with Gate edits filled in. diff --git a/scripts/resolvers/contract-done-means.ts b/scripts/resolvers/contract-done-means.ts index f58daf3b2..39b526da7 100644 --- a/scripts/resolvers/contract-done-means.ts +++ b/scripts/resolvers/contract-done-means.ts @@ -9,6 +9,7 @@ import * as fs from 'fs'; import * as path from 'path'; import type { TemplateContext } from './types'; +import { getHostConfig } from '../../hosts/index'; const ROOT = path.resolve(import.meta.dir, '..', '..'); const code = (s: string) => '\u0060' + s + '\u0060'; @@ -31,6 +32,12 @@ const KIND_FIELDS: Record = { agent_dispatch: ['match'], }; +// The resolver that renders the step each kind checks today. A host that suppresses it never shows that step, so its items are left out of the list there. +const BACKED_BY: Partial> = { + selected_specialists_dispatched: 'REVIEW_ARMY', + agent_dispatch: 'ADVERSARIAL_STEP', +}; + const PLACEHOLDER = /^\[.*\]$|^<.*>$|^\.\.\.$|^(todo|tbd|n\/a)$/i; function fail(where: string, problem: string): never { @@ -101,12 +108,13 @@ function plainWords(item: ContractItem): string { switch (item.kind) { case 'report_labels': return 'a ' + code(item.heading) + ' with ' + list(item.labels) + ' filled in'; case 'check_after_last_edit': return 'a test or repro command run after your last edit'; - case 'selected_specialists_dispatched': return 'the specialist selection printed (' + code(item.marker) + ' ...), then one Agent dispatch per specialist it names'; + case 'selected_specialists_dispatched': return 'the specialist selection printed (' + code(item.marker) + ' ...), then one Agent dispatch per specialist it names (none when a small diff skips them)'; case 'agent_dispatch': return 'an Agent dispatch for the ' + item.id.replace(/-/g, ' ') + ', its prompt including "' + item.match + '"'; } } export function renderDoneMeans(contract: SkillContract): string { + if (contract.items.length === 0) return ''; return [ '## Done means', '', @@ -121,6 +129,7 @@ export function renderDoneMeans(contract: SkillContract): string { export function generateContractDoneMeans(ctx: TemplateContext): string { const where = path.relative(ROOT, ctx.tmplPath).split(path.sep).join('/'); - const contract = contractFromTemplate(fs.readFileSync(ctx.tmplPath, 'utf-8'), where, { required: true }); - return renderDoneMeans(contract!); + const contract = contractFromTemplate(fs.readFileSync(ctx.tmplPath, 'utf-8'), where, { required: true })!; + const suppressed = new Set(getHostConfig(ctx.host).suppressedResolvers ?? []); + return renderDoneMeans({ ...contract, items: contract.items.filter(item => !suppressed.has(BACKED_BY[item.kind] ?? '')) }); } diff --git a/test/fixtures/skill-md-size-allowlist.json b/test/fixtures/skill-md-size-allowlist.json index 24a129a49..4dc0e9e98 100644 --- a/test/fixtures/skill-md-size-allowlist.json +++ b/test/fixtures/skill-md-size-allowlist.json @@ -6,52 +6,52 @@ "agy/gstack-plan-design-review": 109107, "agy/gstack-plan-devex-review": 114948, "agy/gstack-plan-eng-review": 143581, - "agy/gstack-review": 104809, + "agy/gstack-review": 104663, "codex/gstack-autoplan": 108400, "codex/gstack-plan-design-review": 110464, "codex/gstack-plan-devex-review": 116456, "codex/gstack-plan-eng-review": 144746, - "codex/gstack-review": 104332, + "codex/gstack-review": 104186, "copilot/gstack-autoplan": 109100, "copilot/gstack-plan-design-review": 109373, "copilot/gstack-plan-devex-review": 114410, "copilot/gstack-plan-eng-review": 143746, - "copilot/gstack-review": 104390, + "copilot/gstack-review": 104244, "cursor/gstack-autoplan": 108332, "cursor/gstack-plan-design-review": 107915, "cursor/gstack-plan-devex-review": 113680, "cursor/gstack-plan-eng-review": 142336, - "cursor/gstack-review": 122680, + "cursor/gstack-review": 122716, "factory/gstack-autoplan": 110610, "factory/gstack-plan-design-review": 111013, "factory/gstack-plan-devex-review": 117043, "factory/gstack-plan-eng-review": 145444, - "factory/gstack-review": 126453, + "factory/gstack-review": 126489, "gbrain/gstack-autoplan": 108739, "gbrain/gstack-plan-eng-review": 124963, "gemini/gstack-autoplan": 108890, "gemini/gstack-plan-design-review": 109201, "gemini/gstack-plan-devex-review": 114238, "gemini/gstack-plan-eng-review": 143579, - "gemini/gstack-review": 104212, + "gemini/gstack-review": 104066, "hermes/gstack-autoplan": 108742, "hermes/gstack-plan-eng-review": 124999, "kiro/gstack-autoplan": 108106, "kiro/gstack-plan-design-review": 107761, "kiro/gstack-plan-devex-review": 113532, "kiro/gstack-plan-eng-review": 142198, - "kiro/gstack-review": 122504, + "kiro/gstack-review": 122540, "openclaw/gstack-autoplan": 108562, "openclaw/gstack-plan-eng-review": 123478, "opencode/gstack-autoplan": 109081, "opencode/gstack-plan-design-review": 109229, "opencode/gstack-plan-devex-review": 114288, "opencode/gstack-plan-eng-review": 143605, - "opencode/gstack-review": 123302, + "opencode/gstack-review": 123338, "slate/gstack-autoplan": 108219, "slate/gstack-plan-design-review": 107838, "slate/gstack-plan-devex-review": 113606, "slate/gstack-plan-eng-review": 142267, - "slate/gstack-review": 122592 + "slate/gstack-review": 122628 } } diff --git a/test/helpers/carve-guards.ts b/test/helpers/carve-guards.ts index c4afb431a..795dbdd44 100644 --- a/test/helpers/carve-guards.ts +++ b/test/helpers/carve-guards.ts @@ -490,10 +490,10 @@ do not launch the downstream skill or open a browser.`, gateAfterStop: undefined, // operational multi-STOP skill, like ship }, behavioral: 'plan', - maxSkeletonBytes: 59_913, // GST-1088 lane B (raised from 59,073 to measured, no headroom): the contract: frontmatter block and the F1 "Done means" list rendered from it at the top of the body, plus widened triggers; F1 is a gate-rev-2 recommendation on the plan Garry approved with "do it" on gate rev 1 (2026-11-03), contingent on his gate-rev-2 answer, and lane C's diet is expected to bring /review back under 59,073; measured 59,913. Before: GSTA-96 hoist (2026-10-26): lowered from 80_900 by the shared preamble that moved to sections/preamble.md, keeping the pre-hoist headroom of 319 bytes; measured 58,754. Before: Shared-code identity/skip/action rules + critical-severity validation; measured 74,493 (2026-09-17). + v1.91.12.0 merge of #2999 (review clarity repairs: await reads, research alongside dispatch, /review deadline and setup authority, findings sources) with #3002 (guarded state-root lines, plan-check checkpoints); each fit alone; measured 74,881 (2026-10-01). + Oct 8 wave: shared voice directive bans "load-bearing" and adds the reply-language rule, offset by tightening the bounded closer (net +17 bytes per tier-2+ skill); measured 74,884 (2026-10-08). + gate integrity (HONEST_WORK_GATE_INTEGRITY W2): Step 3.5 Diff scans with the shared GATE_SCAN_BLOCK, the Step 4 carry, the four-option `review-gate-disposition` question in 5c, gate record fields in 5.8 and the Gate edits report line; measured 80,054 (2026-10-10). + Oct 10 multi-agent wave PR D2 (sideways review): Step 4's enum-only paragraph becomes the sideways sweep with the forced-CRITICAL list, plus the flake-fix number sentence in the Gate Integrity paragraph; measured skeleton 80,812 (2026-10-10). + maxSkeletonBytes: 59_949, // GST-1088 lane B (raised from 59,073 to measured, no headroom): the contract: frontmatter block and the F1 "Done means" list rendered from it at the top of the body, plus widened triggers; F1 is a gate-rev-2 recommendation on the plan Garry approved with "do it" on gate rev 1 (2026-11-03), contingent on his gate-rev-2 answer, and lane C's diet is expected to bring /review back under 59,073; measured 59,949. Before: GSTA-96 hoist (2026-10-26): lowered from 80_900 by the shared preamble that moved to sections/preamble.md, keeping the pre-hoist headroom of 319 bytes; measured 58,754. Before: Shared-code identity/skip/action rules + critical-severity validation; measured 74,493 (2026-09-17). + v1.91.12.0 merge of #2999 (review clarity repairs: await reads, research alongside dispatch, /review deadline and setup authority, findings sources) with #3002 (guarded state-root lines, plan-check checkpoints); each fit alone; measured 74,881 (2026-10-01). + Oct 8 wave: shared voice directive bans "load-bearing" and adds the reply-language rule, offset by tightening the bounded closer (net +17 bytes per tier-2+ skill); measured 74,884 (2026-10-08). + gate integrity (HONEST_WORK_GATE_INTEGRITY W2): Step 3.5 Diff scans with the shared GATE_SCAN_BLOCK, the Step 4 carry, the four-option `review-gate-disposition` question in 5c, gate record fields in 5.8 and the Gate edits report line; measured 80,054 (2026-10-10). + Oct 10 multi-agent wave PR D2 (sideways review): Step 4's enum-only paragraph becomes the sideways sweep with the forced-CRITICAL list, plus the flake-fix number sentence in the Gate Integrity paragraph; measured skeleton 80,812 (2026-10-10). minUnionBytes: 89_000, // Phase 4 wave 1; measured union 93,357 mustContain: ['confidence', 'P1', 'P2', 'Review Army', 'adversarial'], - maxSizeRatio: 1.3274, // + GST-1088 lane B (raised from 1.3180 to measured, no headroom): the contract: frontmatter block and the F1 "Done means" list, contingent on Garry's gate-rev-2 answer; lane C's diet is expected to bring /review back under 1.3180; measured 144,046 / 108,523 = 1.3274. Before: + GSTA-167 variant G part 1 (plan approved by Garry 2026-10-30, autoplan-20261030161754 "Do it", which names about +0.3k prompt tokens per voice): the native adversarial and outside-voice prompts carry the blocking floor and the Severity and Blocks lines, +537 bytes; measured 1.3180 (2026-10-30). Before: 1.3136. GSTA-96 hoist merged with GSTA-92 phase 2 (2026-10-26): both owner-approved raises together, measured 142,550 / 108,523 = 1.3135. Before: Shared-code feature + critical-severity validation: 128,042 union bytes / 108,523 baseline = 1.1799; preserves content floors. + v1.91.12.0 merge of #2999 (above, plus plan-completion fallback intent and specialist checklist-by-path) with #3002; measured 1.1843 (2026-10-01). + v1.91.19.0 wave: verdict-form Codex callers with sandbox preflight (INV-1/B1), P0/P1 structured gate (B1b), bound-plan discovery and data-not-instructions guard (F1/D3), visible LEARNINGS failures (B5); measured 129,941 / 108,523 = 1.1974 (2026-10-03). + v1.91.18.0 (#3023) merged into the v1.91.19.0 wave; measured 130,025 / 108,523 = 1.1981 (2026-10-04). + Oct 6 fix wave: quota_exhausted preflight branch (#3051); measured 130,359 / 108,523 = 1.2012 (2026-10-06). + Oct 6 fix wave A3/CEO-12: free text reaches commands only through agent-written mktemp files (mktemp block, write rule, path rebuild and refusal at each posting, prompt and brief site); measured 131,314 / 108,523 = 1.2100 (2026-10-06). Combined Oct 6 wave head (A and B): measured 131,966 / 108,523 = 1.2160 (2026-10-06). + Oct 7 wave B3: the Codex probe runs as a command (select-model capture lines, check-sandbox/probe-model/run-with-timeout subcommands) instead of being sourced; measured 132,199 / 108,523 = 1.2182 (2026-10-07). + Oct 7 wave CEO-28/DX-11: review prompts ask for a severity per finding and design proposals use the proposal gate; measured 132,258 / 108,523 = 1.2187 (2026-10-07). + Oct 7 wave guard (CEO-20): conditional `run_in_background` request plus the background-run recovery at each foreground dispatch site; measured 132,233 / 108,523 = 1.2185 (2026-10-07). Merged with the guard area: measured 132,522 / 108,523 = 1.2211 (2026-10-07). + Oct 8 wave: the defensive-security framing and fixture summary mode reach the security specialist and the Red Team (stated once in the dispatch, referenced by the Red Team); measured 133,902 / 108,523 = 1.2339 (2026-10-08). + gate integrity W2 (above); measured 1.282 (2026-10-10). + Oct 10 multi-agent wave PR D2 (sideways review): sideways sweep and flake-fix number in Step 4; `Diff read` and `Undeclared behavior changes` in the review-mode Scope Check (plan-completion section); measured 140,497 / 108,523 = 1.2946 (2026-10-10). + harness lane E (2026-10-26): the lead-judgment stage in the finding merge and the `cause` anchor in the specialist prompts, shared with /ship through scripts/resolvers/review-army.ts; measured 141,949 / 108,523 = 1.3080. + maxSizeRatio: 1.3277, // + GST-1088 lane B (raised from 1.3180 to measured, no headroom): the contract: frontmatter block and the F1 "Done means" list, contingent on Garry's gate-rev-2 answer; lane C's diet is expected to bring /review back under 1.3180; measured 144,082 / 108,523 = 1.3277. Before: + GSTA-167 variant G part 1 (plan approved by Garry 2026-10-30, autoplan-20261030161754 "Do it", which names about +0.3k prompt tokens per voice): the native adversarial and outside-voice prompts carry the blocking floor and the Severity and Blocks lines, +537 bytes; measured 1.3180 (2026-10-30). Before: 1.3136. GSTA-96 hoist merged with GSTA-92 phase 2 (2026-10-26): both owner-approved raises together, measured 142,550 / 108,523 = 1.3135. Before: Shared-code feature + critical-severity validation: 128,042 union bytes / 108,523 baseline = 1.1799; preserves content floors. + v1.91.12.0 merge of #2999 (above, plus plan-completion fallback intent and specialist checklist-by-path) with #3002; measured 1.1843 (2026-10-01). + v1.91.19.0 wave: verdict-form Codex callers with sandbox preflight (INV-1/B1), P0/P1 structured gate (B1b), bound-plan discovery and data-not-instructions guard (F1/D3), visible LEARNINGS failures (B5); measured 129,941 / 108,523 = 1.1974 (2026-10-03). + v1.91.18.0 (#3023) merged into the v1.91.19.0 wave; measured 130,025 / 108,523 = 1.1981 (2026-10-04). + Oct 6 fix wave: quota_exhausted preflight branch (#3051); measured 130,359 / 108,523 = 1.2012 (2026-10-06). + Oct 6 fix wave A3/CEO-12: free text reaches commands only through agent-written mktemp files (mktemp block, write rule, path rebuild and refusal at each posting, prompt and brief site); measured 131,314 / 108,523 = 1.2100 (2026-10-06). Combined Oct 6 wave head (A and B): measured 131,966 / 108,523 = 1.2160 (2026-10-06). + Oct 7 wave B3: the Codex probe runs as a command (select-model capture lines, check-sandbox/probe-model/run-with-timeout subcommands) instead of being sourced; measured 132,199 / 108,523 = 1.2182 (2026-10-07). + Oct 7 wave CEO-28/DX-11: review prompts ask for a severity per finding and design proposals use the proposal gate; measured 132,258 / 108,523 = 1.2187 (2026-10-07). + Oct 7 wave guard (CEO-20): conditional `run_in_background` request plus the background-run recovery at each foreground dispatch site; measured 132,233 / 108,523 = 1.2185 (2026-10-07). Merged with the guard area: measured 132,522 / 108,523 = 1.2211 (2026-10-07). + Oct 8 wave: the defensive-security framing and fixture summary mode reach the security specialist and the Red Team (stated once in the dispatch, referenced by the Red Team); measured 133,902 / 108,523 = 1.2339 (2026-10-08). + gate integrity W2 (above); measured 1.282 (2026-10-10). + Oct 10 multi-agent wave PR D2 (sideways review): sideways sweep and flake-fix number in Step 4; `Diff read` and `Undeclared behavior changes` in the review-mode Scope Check (plan-completion section); measured 140,497 / 108,523 = 1.2946 (2026-10-10). + harness lane E (2026-10-26): the lead-judgment stage in the finding merge and the `cause` anchor in the specialist prompts, shared with /ship through scripts/resolvers/review-army.ts; measured 141,949 / 108,523 = 1.3080. }, codex: { skill: 'codex', diff --git a/test/skill-descriptions.test.ts b/test/skill-descriptions.test.ts index 6000dd113..97c0bd7b8 100644 --- a/test/skill-descriptions.test.ts +++ b/test/skill-descriptions.test.ts @@ -2,7 +2,8 @@ import { describe, expect, test } from 'bun:test'; import { existsSync, readdirSync, readFileSync } from 'node:fs'; import { join } from 'node:path'; import { sentences } from '../lib/skill-triggers'; -import { contractFromTemplate, parseSkillContract, renderDoneMeans } from '../scripts/resolvers/contract-done-means'; +import { contractFromTemplate, generateContractDoneMeans, parseSkillContract, renderDoneMeans } from '../scripts/resolvers/contract-done-means'; +import { HOST_PATHS } from '../scripts/resolvers/types'; /** * GST-1088 discovery targets: each says when to use it in the part of its @@ -158,7 +159,7 @@ describe('contract-done-means renderer', () => { expect(out).toContain('## Done means'); expect(out).toContain('- ' + BT + 'debug-report' + BT + ': a ' + BT + 'DEBUG REPORT' + BT + ' with Symptom, Root cause and Status filled in.'); expect(out).toContain('- ' + BT + 'check-after-last-edit' + BT + ': a test or repro command run after your last edit.'); - expect(out).toContain('- ' + BT + 'specialists-dispatched' + BT + ': the specialist selection printed (' + BT + 'Dispatching N specialists:' + BT + ' ...), then one Agent dispatch per specialist it names.'); + expect(out).toContain('- ' + BT + 'specialists-dispatched' + BT + ': the specialist selection printed (' + BT + 'Dispatching N specialists:' + BT + ' ...), then one Agent dispatch per specialist it names (none when a small diff skips them).'); expect(out).toContain('- ' + BT + 'adversarial-pass' + BT + ': an Agent dispatch for the adversarial pass, its prompt including "chaos engineer".'); }); @@ -195,6 +196,15 @@ describe('contract-done-means renderer', () => { }); } + test('a host that suppresses the step an item checks leaves that item out', () => { + const render = (host: string) => generateContractDoneMeans({ skillName: 'review', tmplPath: join(ROOT, 'review', 'SKILL.md.tmpl'), host, paths: HOST_PATHS[host] } as Parameters[0]); + const ids = (host: string) => [...render(host).matchAll(/^- \u0060([a-z-]+)\u0060:/gm)].map(m => m[1]); + expect(ids('claude')).toEqual(['specialists-dispatched', 'adversarial-pass', 'review-report']); + expect(ids('codex')).toEqual(['adversarial-pass', 'review-report']); + expect(ids('hermes')).toEqual(['review-report']); + expect(renderDoneMeans({ version: 1, items: [] })).toBe(''); + }); + test('a placeholder with no contract block fails the render', () => { expect(() => contractFromTemplate('---\nname: demo\n---\n{{CONTRACT_DONE_MEANS}}\n', 'demo/SKILL.md.tmpl', { required: true })) .toThrow(/CONTRACT_INVALID demo\/SKILL.md.tmpl: \{\{CONTRACT_DONE_MEANS\}\} needs a contract block/);