一键导入
maestro-session-seal
Seal current session with knowledge extraction and DAG progression
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Seal current session with knowledge extraction and DAG progression
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Manifest-driven knowledge asset generator — converts any structured package into maestro knowhow + spec entries with ref linking. Triggers on "codify-to-knowhow", "style to knowhow", "知识固化".
Check workflow delegation prompts against agent role definitions for content separation violations. Detects conflicts, duplication, boundary leaks, and missing contracts. Triggers on "check delegation", "delegation conflict", "prompt vs role check".
Adversarial review of code quality findings. Challenges insights with counter-evidence, verifies claims against source code, and produces structured verdicts. Triggers on 'insight-challenge', 'challenge finding', '审查发现'.
Quick execution for small tasks (≤1-2 files, no artifact handoff) — minimal run lifecycle (create + complete only) with evidence recording. Can read/write/run any tool, but scoped to tasks completable in a few actions. Not for multi-step workflows or tasks needing downstream gates
Create or sync session worktree for parallel dev
Manage editing boundary restrictions
| name | maestro-session-seal |
| disable-model-invocation | true |
| description | Seal current session with knowledge extraction and DAG progression |
| argument-hint | [--session <session_id>] [-y] [--skip-knowledge] |
| allowed-tools | ["Bash","Edit","Glob","Grep","Read","Write","followup_task","interrupt_agent","list_agents","request_user_input","send_message","spawn_agent","spawn_agents_on_csv","wait_agent"] |
| session-mode | run |
| contract | {"discovery":"self-described","consumes":[],"produces":[]} |
| version | 0.5.55 |
<required_reading>
@/.maestro/workflows/run-mode.md
@/.maestro/workflows/codex-run-mode.md
</required_reading>
Replaces the deprecated maestro-milestone-complete with session-level semantics and integrated knowledge capture.
Flags:
| Flag | Effect | Default |
|---|---|---|
--session <id> | Target session (slug or full ID) | active_session_id |
-y / --yes | Auto mode — skip confirmations | false |
--skip-knowledge | Skip knowledge extraction step | false |
--session flag or active_session_idsession.json — verify status is running or pausedreview step first")Skip if --skip-knowledge. Otherwise:
status: accepted from runs/*/run.json.handoff.decisions[] → spec candidatesrequest_user_input:
question: "以下知识候选项值得持久化吗?"
options:
- "全部保存" (save all candidates as specs/knowhow)
- "逐个选择" (review each candidate)
- "跳过" (no knowledge extraction)
/maestro-spec add .../maestro-manage knowledge capture ...session.json.lifecycle.promoted[](前缀区分 spec:/knowhow:)maestro run seal-session {session_id}session.json.lifecycle.sealed_at and seal_summarystate.json.sessions[].status to sealedstate.json.sessions[] — find sessions that became dep-ready (all depends_on sealed)question: "Session {slug} 已 sealed。推荐激活下一个 session: {next-slug},是否确认?"
options:
- "激活推荐 session"
- "选择其他 session"
- "暂不激活"
active_session_id to selected session| Condition | Suggestion |
|---|---|
| Next session activated | step analyze (maestro run prepare --platform codex analyze + maestro run create analyze --session {next-slug} --intent "{goal}") |
| DAG complete (all sealed) | /maestro-manage status |
| Knowledge review needed | /maestro-manage knowledge audit |
<error_codes>
| Code | Severity | Condition | Recovery |
|---|---|---|---|
| E001 | error | Session not found | Check state.json.sessions[] |
| E002 | error | Session already sealed | Nothing to do |
| E003 | error | Active runs exist | Complete or seal pending runs first |
| E004 | error | Critical gates failed | Run verify/review to resolve |
| W001 | warning | No knowledge candidates found | Proceed to seal |
| </error_codes> |
<success_criteria>
session.json.lifecycle.sealed_at written)state.json.sessions[].status updated to sealed