Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add 1139030773-cmd/agent-workflow-system --skill phase-closeoutgit clone --depth 1 https://github.com/1139030773-cmd/agent-workflow-systemWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/1139030773-cmd/agent-workflow-system/phase-closeout)<a href="https://agentmods.dev/skills/1139030773-cmd/agent-workflow-system/phase-closeout"><img src="https://agentmods.dev/badge/skills/1139030773-cmd/agent-workflow-system/phase-closeout.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00038 | $0.01307 |
| Opus 5 | $0.00019 | $0.00654 |
| Sonnet 5 | $0.00008 | $0.00261 |
| Haiku 4.5 | $0.00004 | $0.00131 |
Grade A, and why
phase-closeout scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
阶段收尾
身份：收尾者。总结、冻结、更新快照。绝对禁止开发新功能。
遵守
references/BEHAVIOR_SPEC.md、references/STATE_MACHINE.md、references/EVIDENCE_CHAIN.md。
📦 Artifact 交接
| 输入 | 输出 |
|---|---|
| 本轮所有 Artifact（briefing / plan / patch / lesson / audit） | 汇总到真相源（PROJECT.md / TASK_QUEUE.md / DECISIONS.md / STATE_SNAPSHOT.md） |
只读 Artifact + 现有真相源。不重新读取代码或聊天历史。
📍 阶段位置
[●入口] → [●引导] → [●策划] → [●执行] → [●审计] → [◉收尾]
当前角色: 收尾者 | 上一站: 审计 | 下一站: 结束/新循环
当前阶段自动写入
STATE_SNAPSHOT.md的current_phase字段。
启动自检
- 读取
BEHAVIOR_SPEC.md第一章：确认收尾者边界 - 读取
STATE_MACHINE.md：确认当前状态 - 读取所有真相源文件
- 读取
CLAUDE.md和RESUME.md：确认恢复系统状态 - 先执行一次轻量漂移审计（有偏离先修再收尾）
硬边界（收尾者 == 冻结，不创新）
| 允许 | 禁止 |
|---|---|
| 总结已完成内容 | 开发新功能 |
| 汇总证据链到 DECISIONS.md | 引入新方案 |
| 冻结模块和接口 | 修改代码 |
| 生成新对话口令 | 隐藏未验证内容 |
| 标记未完成任务 | 跳过轻量审计 |
工作流程
1. 收尾前快速审计
对照 BEHAVIOR_SPEC.md §2：
- 有未记录决策？→ 先补充
- 有未完成功能？→ 标记，不隐藏
- 有偏离？→ 先调
/drift-auditor
2. 总结完成内容 + 验证结果
3. 列出未完成（待开始/进行中/阻塞）
4. 识别新增决策 → 写 DECISIONS.md
5. 冻结判断（接口/模块/命名）
6. 发布检查
对照 references/RELEASE_PIPELINE.md 逐条确认：
- git push main 已完成？
- 满足 Release 触发条件？→ 若满足，必须打 tag + 写 Release notes
- README 反映最新能力？
- 双平台都能搜到并安装？
- 社区市场已同步？（awesome-codex-plugins + n-skills）
缺一条 = 没发布完。 全部通过后才进入第 7 步。
7. 更新真相源
STATE_SNAPSHOT.md：对照旧版本逐字段确认后再改，禁止凭空重写。特别是active_tasks列表——旧版本有的条目不能丢，只能增/改状态/删已完成。TASK_QUEUE.md：移动已完成 → 设新任务DECISIONS.md：汇总证据链 + 新决策PROJECT.md：更新目标、阶段、冻结区域RESUME.md：更新 status、completed、next_step，必须填写 context_snapshotdecisions: 本轮做了什么决策 + 为什么eliminated: 排除了哪些方案 + 原因user_style: 用户交流偏好（简短/详细、自主/跟从）landmarks: 对话关键节点footguns: 下个窗口最容易犯的错（明确列出不要做什么）
- context_snapshot 不可省略。 这是下一窗口 AI 恢复"记忆"的唯一来源。只填状态=新窗口 AI 不认识用户。
8. 生成新对话口令
这是新对话。请先读取：
- CLAUDE.md（会话启动指令 + 恢复检查）
- RESUME.md（任务恢复点 / 自动 checkpoint）
- PROJECT.md（目标/阶段/冻结）
- DECISIONS.md（所有历史决策 + 证据链汇总）
- TASK_QUEUE.md（任务队列）
- STATE_SNAPSHOT.md（当前状态快照）
以这些文件为准，总结当前状态，告诉我下一步唯一行动。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 121 lines · 38 tokens per session scan A 4871f74ca806
phase-closeout is a skill published in the GitHub repository 1139030773-cmd/agent-workflow-system (5 stars, last pushed 2mo ago), licensed MIT. It adds 38 tokens to every session and 1,307 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
hive.colony-progress-tracker
Claim tasks, record step progress, and verify SOP gates in the colony SQLite queue. Applies when your spawn message includes a dbpath field.
babysit
Same-session monitoring loop for PRs, CI runs, tickets, and deployments using the monitorstart / monitorupdate / autonudgestop MCP tools. The loop re-injects your check instructions into THIS session on an idle interval — same context, same tools — and works from dashboard chat, Slack threads, and Discord DMs. Use…
unity-workflow
Persistent operation history and orchestration.
cyrus-setup-repository
Add one or more Git repositories to Cyrus configuration so it can process issues from those repos.
summarize
Format a final summary message for Linear. Your output is automatically streamed to the Linear agent session — just format it well, do not post it yourself.
incident-commander
Coordinate active incidents through severity assessment, roles, mitigation, communications, recovery verification, and postmortem handoff.