Debug workers sharing mutable state in the research-critique coordinator

Do not start with a speculative fix for the research-critique coordinator. Force the failure on purpose, save the before output, then change one cause at a time. Retries are allowed only for transient conditions—not for bad input that will fail forever on multi-agent-in-code.

for (let attempt = 0; attempt <= maxRetries; attempt++) {
  try { return await operation(signal); }
  catch (error) {
    const failure = classify(error);
    if (!failure.retryable || attempt === maxRetries) throw failure;
    await sleep(jitteredBackoff(attempt), signal);
  }
}
throw new Error("unreachable");

Expected evidence: Research returns two sourced claims; critique rejects one unsupported claim; synthesis keeps one.. If you cannot reproduce on demand, you do not yet control the failure mode for multi-agent-in-code.

After repair, rerun the exact reproduction command. Keep the failing fixture as a regression seed for the observability page. For the research+critique coordinator, remember the claim you are restoring: critique can veto an unsupported research draft; coordinator never exceeds worker budgets.

Store a one-command reproduction for: workers sharing mutable state, or coordinator ignoring critique veto. The command should use two workers with separate step budgets or a minimal mutant of it. Paste the failing output into notes/failure-before.txt (or your shell scrollback as copied text). After the fix, paste notes/failure-after.txt and keep both.

Retries belong only on transient faults. If the failure is bad input, a bad allowlist, or a logic bug in the research-critique coordinator, retrying will amplify cost without repairing trust around delegate research and critique to two bounded workers, then merge under a coordinator policy.

Classify the failure as prevent, detect, contain, or recover—using this lab’s language, not a generic poster. For multi-agent-in-code, the first fix should usually be detect+prevent at the boundary, because workers sharing mutable state, or coordinator ignoring critique veto is cheaper to stop early than to explain in production prose.

At the debugging stage for multi-agent-in-code, the job is narrower than finishing a product demo. You are creating one progressive evidence piece about two workers with separate step budgets that later pages inherit without redefining success. Keep that fixture small enough to inspect by hand, keep outputs copy-pasteable as text, and refuse to narrate this baseline as if it were a production SLA: single agent without critique pass.

For this page specifically, success looks like before/after evidence for the characteristic failure while still centering the user decision to delegate research and critique to two bounded workers, then merge under a coordinator policy. If you cannot point to a file, command, or assertion that proves that for the research-critique coordinator, stay on this page instead of advancing.

Glossary: tool · Glossary: structured output · Cheatsheet: production ops signals

Previous · Next