v1.90.2.0 perf: halve local free-suite time and preserve coverage (#2972)

* perf: remove repeated test work and preserve AUQ execution budgets

* fix: validate native evaluation fixture evidence at its actual boundaries

* fix: clarify deployment approval and recovery state transitions

* chore: document coverage and release v1.90.2.0

* test: preserve Windows scheduling and native no-change consent

* test: recognize verified reads through fixture symlinks

* test: isolate alias-name installation from runtime assets
This commit is contained in:
Garry Tan
2026-09-25 13:27:07 -04:00
committed by GitHub
parent a84b0b5b6d
commit 7b534d3e90
70 changed files with 4050 additions and 999 deletions
+2 -1
View File
@@ -110,12 +110,13 @@ let stage='startup';process.stdin.setRawMode?.(true);process.stdin.on('data',dat
native('user',[{type:'tool_result',tool_use_id:'finding',content:'Answered'}],{toolUseResult:{answers:{[q.question]:'Fix'}}});
process.stdout.write('\x1b[2J\x1b[HDone.\r\n');
});process.on('SIGINT',()=>process.exit(0));process.stdin.resume();
process.stdout.write('PTY_READY:'+item.events+'\x1b[2J\x1b[H');
`);
fs.chmodSync(fake, 0o755);
const runner = pathToFileURL(path.join(import.meta.dir, 'helpers/claude-pty-runner.ts')).href;
const picker = pathToFileURL(path.join(import.meta.dir, 'helpers/ceo-approach-pick.ts')).href;
fs.writeFileSync(worker, `import {runPlanSkillCounting} from ${JSON.stringify(runner)};import {pickCeoCountQuestion} from ${JSON.stringify(picker)};
const result=await runPlanSkillCounting({skillName:'plan-ceo-review',slashCommand:'/plan-ceo-review',followUpPrompt:'Review this fixture.',isLastStep0AUQ:()=>true,pickAUQ:pickCeoCountQuestion,defaultPick:1,reviewCountCeiling:1,timeoutMs:26000,env:{TRUNCATED_CASE:${JSON.stringify(JSON.stringify({ events, screen, question: completed.questions[0] }))}}});await Bun.write(${JSON.stringify(output)},JSON.stringify(result));`);
const result=await runPlanSkillCounting({skillName:'plan-ceo-review',slashCommand:'/plan-ceo-review',followUpPrompt:'Review this fixture.',isLastStep0AUQ:()=>true,pickAUQ:pickCeoCountQuestion,defaultPick:1,reviewCountCeiling:1,timeoutMs:26000,startupReadyMarker:${JSON.stringify('PTY_READY:' + events)},env:{TRUNCATED_CASE:${JSON.stringify(JSON.stringify({ events, screen, question: completed.questions[0] }))}}});await Bun.write(${JSON.stringify(output)},JSON.stringify(result));`);
const child = Bun.spawn([process.execPath, worker], { env: { ...process.env, BROWSE_TERMINAL_BINARY: fake, EVALS_HERMETIC: '1' }, stdout: 'pipe', stderr: 'pipe' });
const timer = setTimeout(() => child.kill('SIGKILL'), 30000);
try {