AGENT SKILL

autospec-run

暂无简介

License 许可
MIT
最近更新
2026-11-11
Stars
★ 3
Forks
1

02 资源文件 3

03 SKILL.md 正文

<!-- guardian-block:begin --> guardian body <!-- guardian-block:end -->

Issue start summary [monitor] starting #$ISSUE: [monitor] goal: [monitor] smoke: [monitor] scope:

Immediate next-issue pickup: NO SLEEP after process(ISSUE) fresh queue scan can pick any issue unblocked

BATCH_COMPLETE is a continuation signal, not a terminal state reasoning:deep may reduce a single monitor batch to one issue the orchestrator MUST relaunch automatically until ALL_DONE Never tell the operator to rerun /autospec-run after BATCH_COMPLETE Codex native subagents with explicit agent_type, model, or reasoning_effort MUST use a bounded handoff, not a full-history fork RETRY-LOOP:begin MAX_IMPL_RETRIES directive_context Retry attempt Implementer hit max retries; manual intervention needed auto-implement-active

Full test suite gate AUTOSPEC_FULL_TEST_COMMAND autospec validate If the full suite fails, fix the failure, recommit, rerun the full suite, and repeat Do NOT dispatch LGTM review Do NOT run gh pr merge Record the exact full-suite command and passing output summary

data-scope invariant lens empty optional filters reject unless documented unsupported-filter empty-filter job-only sample-only job+sample

AUTOSPEC_BATCH_SIZE:-1 Fresh-subagent-per-issue (canonical Phase 4 path, formerly single-agent absorbed-discipline) The orchestrator NEVER implements in its own context the default is 1 (one issue per subagent) reasoning:deep` force-to-1 rule is retained

  1. SUCCESS <!-- token-report:begin --> post-token-report.sh <!-- token-report:end -->
  2. FAILURE

_REGEN resolveAutoRegenerate <!-- docs-drift-gate:begin --> doc-orchestrator.mjs docs: regeneration skipped (auto_regenerate=false) <!-- docs-drift-gate:end -->

04 同仓库更多 Skills