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 AbyssCN/oh-my-dag --skill retrogit clone --depth 1 https://github.com/AbyssCN/oh-my-dagWrote 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/abysscn/oh-my-dag/retro)<a href="https://agentmods.dev/skills/abysscn/oh-my-dag/retro"><img src="https://agentmods.dev/badge/skills/abysscn/oh-my-dag/retro/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/abysscn/oh-my-dag/retro"><img src="https://agentmods.dev/badge/skills/abysscn/oh-my-dag/retro.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00119 | $0.01218 |
| Opus 5.5 | $0.00048 | $0.00487 |
| Sonnet 5.5 | $0.00024 | $0.00244 |
| Haiku 4.5 | $0.00012 | $0.00122 |
Grade A, and why
retro 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 5d 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.
What it actually says
/retro — 批后复盘:把重复的坑分流成闸或审查标准
契约:docs/plan/2026-10-08-mcode-loop-retro-and-spec-axis.md(oh-my-dag)§3。分类法来自 mattpocock/skills 的 retro。
时机:一批 omdl run 合并进 main 之后跑一次。不自动改闸、不自动改模板(D7)—— 产出是提案,owner 批了才落地。
步骤
- 汇总(零模型):
omd-lite retro --since <批次开始日 YYYY-MM-DD>(可再跟 label 只看某几个 run)。 读末尾的「聚合」:首败闸 / 执行器失败 · 越界文件 · 执行器失败 · not-run / 中断原因 · 多轮占比。聚合只列 ≥ 2 个 run 的模式(D6);单次的只记不提。执行器挂了(<码>)是执行器 / 供应商 / 调度的问题(限流、并发锁agent_name_conflict…:lock、CONTENT_REVIEW_BLOCKED、spawn 失败), 去处通常是调度规则或驱动,不是任务书;执行体没交付(没有改动)才要回头读任务书与执行体的收尾报告。- 越界文件 ≥ 2 次 = 「允许改动」漏列候选(典型:登记处 —— i18n messages、规则注册表、index 导出)。
- 清 worktree(预览):
omd-lite gc(只打印路径 · 原因 · 大小)。把可清清单与总大小写进产出;owner 批了再omd-lite gc --apply。 不对正在跑的批次跑--apply;gc 自己也会跳过正在跑的 run、有未提交改动的、有被忽略产物的 worktree。 - 补两路人工信号(机器数不到的):
- 本批 Aalto 返工:
git log main --since <批次开始日> --grep '<run 名>'里 merge 之后带 run 名的修正提交,读改了什么; - 本批 review(
/omd-review,omdl run 一律带sdd: <任务书>)里被 accept 的项。
- 本批 Aalto 返工:
- 定位:每个重复模式先按七类归因 —— Navigation(找不到该改的地方)· Automated checks(闸缺 / 闸错)· Coding standards · AGENTS.md 瘦身(上下文太长被忽略)· Tool economy(工具用法浪费轮次)· No-ops(空转、没交付)· Information access(读不到该读的)。
- 分流(D5:机械可判 → 确定性闸,没有例外;要判断 → 审查标准给 reviewer,不加给执行方):
- 闸:plana
scripts/dev/gate-*.sh加一道,或 omd-lite 越界闸加预检。写出具体命令和它在本批哪几个 run 上会红。 - 审查标准:追加到 reviewer 的
--extra焦点清单(planadocs/process/下一份文件,首次 retro 时建)。 - 任务书模板 / executor-brief:只放导航指针和「Out of scope」,不放规范条文(执行方上下文压力最大,条文塞给它不生效)。
- 不处理:写明为什么(一次性、已修、成本高于收益)。
- 闸:plana
- 产出:在 plana
docs/process/omd-workhorse.md追加## <日期> · retro一节:聚合表原样 + gc 预览一行 + 提案表 (模式 · 证据 run · 七类 · 去处 · 具体改法),交 owner 批。批之前不改闸、不改模板、不跑gc --apply。
纪律
- 聚合里的数是计数不是结论:先读至少一个证据 run 的
gate-round-<n>.log/mcode-round-<n>.log再下判断(一个在任何干预下都不动的数,通常量的是尺子)。 - 已经修过的坑(例如 harness 自己造的越界)写进「不处理 · 已修」,注明修它的提交,别再提一遍。
- 单次出现的模式不进提案表;如果你认为它危险到值得破例,标
?交 owner,不要自己决定。
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.
- 5d ago First seen · 38 lines · 119 tokens per session scan A 1cc8014c5592
retro is a skill published in the GitHub repository AbyssCN/oh-my-dag (43 stars, last pushed 5d ago), licensed MIT. It adds 119 tokens to every session and 1,218 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-10-12.
Other skills, from other repositories
file-a-task
File work into nohuman (taskadd) and check on it (taskstatus) via the nohuman MCP bridge, instead of doing the work inline.
company-employee
The protocol every employee of a PenguinHarness organization follows — read the handbook first, act on [orgtrigger] work runs, schedule tickets from the desk session and do the work in ticket sessions, block instead of idling, ask the board in the all-hands channel before anything heavy, costly, irreversible or…
company-ceo
Run a PenguinHarness organization as its CEO — turn the mission into a ticket tree, hire HR and finance first, partition the shared workspace, schedule the calendar, open a channel per stream, review tickets and report to the board in the all-hands channel.
company-hr
Run HR for a PenguinHarness organization — guarantee every employee has an enabled calendar event, hire and offboard employees (channels included), evaluate and improve them (with the default-rsi-toolkit plugin's agent-optimization skill), and keep the handbook's role conventions current.
company-mirror
Run a PenguinHarness organization that mirrors a real company — one digital twin per real colleague, each twin's desk bound to that colleague's chat bot, twins answering what the handbook already knows and relaying the rest to one another instead of hiring, scheduling and filing tickets.
company-setup
Create a company-mode organization together with the user — collect its id, name, mission, shared workspace, model and CEO budget one question at a time, confirm the whole thing in one summary, then run penguin org create and hand the user over to company mode.