mirror of
https://github.com/garrytan/gstack.git
synced 2026-08-31 18:30:39 +02:00
69 files, both shapes (trailing bun-test budgets and runner timeout/timeoutMs options), ROUND-UP ONLY so nothing that passed can start failing: 75 → JUDGE_MS, 137 → CAPTURE_MS, 74 → CAPTURE_LONG_MS, 9 → PTY_MS, 3 → PTY_LONG_MS. Raw >=60s literal count in the paid scope: 395 → 97, of which 51 are non-timeout noise (fixture dates, run IDs) and 46 are enumerated justified holds (comment-carrying calibrated budgets, poll-loop constants, utility spawn waits, and the seven physical-ceiling 1_500_000 sites). The eval-budgets policy ratchet keeps the residue from regrowing. Known collapse: where an inner runner budget and its enclosing test budget now share a tier, the old stagger is gone — an overrun surfaces as a bun test timeout instead of a graceful runner timeout (diagnosability trade, not a correctness one). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
105 lines
4.3 KiB
TypeScript
105 lines
4.3 KiB
TypeScript
/**
|
|
* T2 — data-driven behavioral section-loading guard (PERIODIC tier, paid, SDK capture).
|
|
*
|
|
* The behavioral proof that a REAL agent actually Reads each carved skill's
|
|
* required sections at runtime — not just that the skeleton structure looks right
|
|
* (that's E2, free, per-PR). One file iterating the canonical CARVE_GUARDS
|
|
* registry (EQ2): registry membership IS the test, so "registered ⇒ asserted" is
|
|
* structural — a carve can't be registered yet behaviorally unguarded.
|
|
*
|
|
* Per codex refined-plan pass:
|
|
* #2 — ONE test() per skill, each with its own timeout + named failure output;
|
|
* a hung claude -p fails only its skill, not the whole file.
|
|
* #3 / D-CODEX(A) — GSTACK_CARVE_SKILL=<name> runs only that skill's case, so
|
|
* the touchfile selector can scope cost to the changed skill; unset runs all.
|
|
* #7 — each case drives the run with the registry's `scenario` (built to force
|
|
* the STOP-Read path) and asserts the required sections were Read.
|
|
*
|
|
* 'external' skills (ship, plan-ceo-review) have bespoke fixtures (git state,
|
|
* Step-0 mode loop) and keep their dedicated tests; E1 asserts those exist.
|
|
*/
|
|
|
|
import { test, expect } from 'bun:test';
|
|
import { CAPTURE_LONG_MS } from './helpers/eval-budgets';
|
|
import { describeE2ETier } from './helpers/e2e-gate';
|
|
import { setupSkillDir, skillFromWorktree, captureSectionReads } from './helpers/auq-sdk-capture';
|
|
import { CARVE_GUARDS } from './helpers/carve-guards';
|
|
|
|
const describeE2E = describeE2ETier('periodic');
|
|
const runId = `carve-section-loading-${process.env.EVALS_RUN_ID ?? 'local'}`;
|
|
const only = process.env.GSTACK_CARVE_SKILL?.trim();
|
|
|
|
// A generic plan fixture for 'plan' behavioral skills (the review family).
|
|
const PLAN_MD = [
|
|
'# Plan: add an in-memory cache layer',
|
|
'',
|
|
'## Context',
|
|
'Reads hit the DB on every request. Add a process-local LRU cache in front of the',
|
|
'read path to cut DB load.',
|
|
'',
|
|
'## Approach',
|
|
'- Wrap the read repository in a cache that stores the last 1000 keys.',
|
|
'- Invalidate on write.',
|
|
'',
|
|
'## Out of scope',
|
|
'Distributed cache, cross-process coherence.',
|
|
'',
|
|
].join('\n');
|
|
|
|
describeE2E('carve behavioral section-loading (periodic, SDK capture)', () => {
|
|
for (const guard of Object.values(CARVE_GUARDS)) {
|
|
// 'external' carves keep their dedicated bespoke tests (E1 verifies those exist).
|
|
if (guard.behavioral === 'external') continue;
|
|
// Cost-scoped selection: when GSTACK_CARVE_SKILL is set, run only that skill.
|
|
if (only && only !== guard.skill) continue;
|
|
|
|
test(
|
|
`${guard.skill}: a real run Reads ${guard.requiredReads.join(', ')}`,
|
|
async () => {
|
|
const { skillMd, sectionsFrom } = skillFromWorktree(guard.skill);
|
|
const fixtures = guard.behavioral === 'plan' ? { 'PLAN.md': PLAN_MD } : {};
|
|
const planDir = setupSkillDir({
|
|
skillName: guard.skill,
|
|
skillMd,
|
|
sectionsFrom,
|
|
fixtures,
|
|
tmpPrefix: `gstack-${guard.skill}-secload-`,
|
|
});
|
|
|
|
const { readSections, reportProduced, output } = await captureSectionReads({
|
|
planDir,
|
|
skillName: guard.skill,
|
|
scenario: guard.scenario,
|
|
reportMarker: /report|review|summary|design doc|handoff/i,
|
|
testName: `${guard.skill} section-loading`,
|
|
runId,
|
|
// 480s, not the helper's 300s default: the heavy full-workflow
|
|
// scenarios (plan-eng-review, office-hours, design-html) satisfy
|
|
// their required section reads inside 60s but need 300-450s of
|
|
// wall clock to finish the report on slower sandboxes — a timeout
|
|
// there reads as a loading failure when the carve invariant held.
|
|
timeout: 480_000,
|
|
});
|
|
|
|
const missing = guard.requiredReads.filter((s) => !readSections.has(s));
|
|
// Named failure output (codex #2): skill + expected + observed.
|
|
expect({
|
|
skill: guard.skill,
|
|
reportProduced,
|
|
expected: guard.requiredReads,
|
|
observed: [...readSections],
|
|
missing,
|
|
}).toEqual({
|
|
skill: guard.skill,
|
|
reportProduced: true,
|
|
expected: guard.requiredReads,
|
|
observed: expect.any(Array),
|
|
missing: [],
|
|
});
|
|
expect(output.trim().length).toBeGreaterThan(200);
|
|
},
|
|
CAPTURE_LONG_MS,
|
|
);
|
|
}
|
|
});
|