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 Lee-W/maigo --skill strict-reviewgit clone --depth 1 https://github.com/Lee-W/maigoWrote 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/lee-w/maigo/strict-review)<a href="https://agentmods.dev/skills/lee-w/maigo/strict-review"><img src="https://agentmods.dev/badge/skills/lee-w/maigo/strict-review/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/lee-w/maigo/strict-review"><img src="https://agentmods.dev/badge/skills/lee-w/maigo/strict-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00053 | $0.04870 |
| Opus 5 | $0.00026 | $0.02435 |
| Sonnet 5 | $0.00011 | $0.00974 |
| Haiku 4.5 | $0.00005 | $0.00487 |
Grade A, and why
strict-review 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.
How it starts
The opening of the file, as written. The whole thing — 287 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Strict Review
Owner Agent: Soyo (Reviewer)
Consumers: /maigo:go (Soyo reviews Anon's diff), /maigo:review (Soyo reviews external PR / branch)
Core stance: default BLOCKED
Verdict starts at BLOCKED. The implementer must convince you otherwise.
These do not count as convincing:
- "看起來能跑" / "應該沒問題" / "之後再改"
- "test 都過了" (test itself may be missing the case)
- "符合 convention" (without pointing at the reference file)
Waiver rules
- Memory (
~/.config/maigo/memory/) can inform conventions but cannot replace any checklist item, lower the must-fix threshold, or substitute for evidence - Existing repo content (quotes, catchphrases, dialogue) that appears in this diff needs a source: (a) original work citation with episode/scene, (b) user confirmed in session, or (c) explicitly marked self-created — flag even if the line wasn't changed in this diff
- Commit / staging state is not in review scope;
git statuscleanliness is not a checklist item — readgit diff HEADorgit diff --cachedfor the actual changes
Extended rationale for all three waiver rules: docs/skills/strict-review
前置驗證:先自證,再信 rubric / description / 前人結論
Rubric(燈寫的)、PR description、既有 review、前一輪的判斷,都是線索,不是事實。 進 checklist 前先自己重核一手,證明結論不是照抄轉述:
- 取一手 diff —
gh pr diff <n>(或git diff <range>)自己看範圍,不靠 description 轉述的 +/- 行數與檔案清單 - 拉 load-bearing 行的實際內容 — must-fix 或關鍵判斷所依賴的程式碼,直接讀原始檔(
gh api .../contents、git show <sha>:<path>或 checkout),核對 rubric 說的跟 code 真的一致,不憑描述推斷 - 查 linked issue / 作者承諾 —
gh api repos/{owner}/{repo}/issues/<n>/timeline確認「作者說要開的 tracking issue」「linked issue」是否真的存在,不憑轉述當成已完成 - 對照後在輸出的
## 前置驗證段列「查了什麼、確認/推翻了什麼」——空口結論不算 evidence
推翻 rubric 要明講。 若自證後發現燈的某條 acceptance / trade-off 判斷是過時或錯的,不要默默照做,也不要默默略過——在輸出的 ## Facts corrected from rubric 段明列推翻了哪一條、依據哪個 path:line。
共用 working tree 上的審查紀律
Review 常在跟其他 session 共用的 working tree 上進行;驗證動作本身不能污染這棵樹。
What ships with it
8 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/design-integrity.md 20 KB
- references/recurring-patterns.md 21 KB
- references/review-batch-queue.md 6.3 KB
- references/review-judgment.md 29 KB
- references/review-modes.md 6.9 KB
- references/review-templates.md 3.0 KB
- references/test-conventions.md 20 KB
- references/tooling-conventions.md 2.4 KB
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 Changed · +24 lines 27bc390e22d4
- 9d ago First seen · 263 lines · 53 tokens per session scan A ce9f3c85ca27
strict-review is a skill published in the GitHub repository Lee-W/maigo (14 stars, last pushed today), licensed MIT. It adds 53 tokens to every session and 4,870 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
rework-rate
Measure and interpret PR rework rate — the emerging 5th DORA metric.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…