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 shimo4228/claude-harness --skill readme-writergit clone --depth 1 https://github.com/shimo4228/claude-harnessWrote 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/shimo4228/claude-harness/readme-writer)<a href="https://agentmods.dev/skills/shimo4228/claude-harness/readme-writer"><img src="https://agentmods.dev/badge/skills/shimo4228/claude-harness/readme-writer/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/shimo4228/claude-harness/readme-writer"><img src="https://agentmods.dev/badge/skills/shimo4228/claude-harness/readme-writer.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 | $0.00303 | $0.05622 |
| Opus 5 | $0.00151 | $0.02811 |
| Sonnet 5 | $0.00061 | $0.01124 |
| Haiku 4.5 | $0.00030 | $0.00562 |
Grade A, and why
readme-writer 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 3d 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 — 255 lines — stays where its author put it; the contents beside it link to each section on GitHub.
readme-writer — Human-Facing README Skill
人間に向けた README を書く・改善するスキル。llms-txt-writer が AI 専用 surface を担うのに対し、
本 skill は 人間 surface の単一正準入口を担う。
重要な事実: README は、grounding 経路（AI 検索 / チャットに repo URL を貼る）で LLM が確実に
前提にできる唯一の surface でもある。そのため README は「人間向けに短く・走査しやすく」しつつ
「LLM が README 一枚だけ読んでもプロジェクトを復元できる小さな情報フロア」を必ず残す。
この両立が本 skill の中心課題（較正と出典は inspiration.md）。
When to Use
- README.md / README.ja.md を新規作成・改善する
- 継ぎ足しで育った README（ADR 参照・姉妹 repo・造語・内部史の密度が上がり、初見で読めない）を根本から作り直す
- GitHub の About（description / topics / homepage） を README と同じ主張に揃える
使わない場面: llms.txt / llms-full.txt / FAQ など AI 専用 doc（→ llms-txt-writer）、
記事・エッセイ（→ writing-ecosystem）、graph.jsonld の設計（→ jsonld-knowledge-graph）。
軸は「人間の ATTENTION × LLM の INFORMATION」
README 最適化の対立軸は「人間向け情報 vs LLM 向け情報」ではない。人間の注意（短く・掴む・走査 できる）× LLM の情報（README だけで復元できる）である。両者は同じ施策に収束するので、 トレードオフでなく設計で両取りする。
- AI 検索 / 引用クローラは llms.txt を実質読まず、
graph.jsonldは直接 fetch では plain text 扱い。 README の情報を「機械層が backstop する」前提で薄くしてはいけない（較正: 「LLM は README しか 読まない」は強すぎる。routed coding agent は llms.txt を on-demand で読む — 詳細はinspiration.md） - two-sided rule: アイデアがどう伝わるかは最適化してよい（見出し階層・entity anchoring・ answer-first の lead）。アイデアが何であるかは曲げない（keyword stuffing・疑問見出し farming・ glossary 投下・主張の歪曲は禁止）。star や引用は成功指標ではない
証拠と判定（code は数える、LLM は判定する）
README 品質は「数えられる事実」と「文脈を読む判断」に分かれる。所有者を分ける:
| 層 | 何を出すか | 所有者 |
|---|---|---|
| 証拠 | 第一画面の行数と新語数、ADR / 他 repo / docs への参照数、造語候補の出現表、<details> の中身、図の直後に prose があるか、内部史・生数値の行、slop 語・em-dash |
code — scripts/readme_evidence.py（JSON。verdict も閾値も exit gate も持たない） |
| 判定 | 第一画面が立つか、段落が読者の問いに答えるか、造語予算、参照が導線か説明の代替か、継ぎ足しの痕跡、論理 | LLM — readme-judge agent（fresh context、集計しない named verdict） |
README に研究値ベースの数値スコアは作らない。「良い入口か」は意味的判断で、同等の決定論的 知見がない。entity density 等の AI surface 指標を人間 README に持ち込むのは anti-pattern。
最小 LLM-read フロア（小さく・非交渉・肥大させない）
規則: README を LLM がそれ一枚だけ読んだ（llms.txt も graph も読まれない）として、 テキストだけでプロジェクトを復元できること。
What ships with it
19 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.
- evals/fixtures/ca-readme-2026-08-19.expected.md 3.0 KB
- evals/fixtures/ca-readme-2026-08-19.md 22 KB
- evals/trigger-eval.json 3.3 KB
- fixtures/sample_clean.md 873 B
- fixtures/sample_issues.md 1.4 KB
- inspiration.md 11 KB
- pyproject.toml 1.6 KB
- references/about.md 5.8 KB
- references/ja-register.md 2.4 KB
- references/readme-judge-checklist.md 9.0 KB
- references/visual.md 5.7 KB
- scripts/__init__.py 0 B runs code
- scripts/readme_evidence.py 26 KB runs code
- tests/__init__.py 0 B runs code
- tests/golden/sample_clean.json 1.9 KB
- tests/golden/sample_issues.json 3.5 KB
- tests/test_evidence_golden.py 1.4 KB runs code
- tests/test_readme_evidence.py 12 KB runs code
- uv.lock 50 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.
- 3d ago Changed d48c005033cf
- 4d ago Changed f064e9077738
- 9d ago First seen · 255 lines · 303 tokens per session scan A 7e5ebbdb09b6
readme-writer is a skill published in the GitHub repository shimo4228/claude-harness (2 stars, last pushed 4d ago), licensed MIT. It adds 303 tokens to every session and 5,622 once invoked, about $0.0015 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
agent-adoption-triage
Five-question triage for routing AI work to the right architecture — script, algorithmic search, LLM workflow, or autonomous agentic loop — before any agent is built. Use when someone says "let's build an agent for X", when deciding whether a use case actually needs an autonomous agent, when reviewing an agent…
job-description-skill
A job-description analysis skill that turns a vacancy listing and a candidate’s background into an HTML report about fit, gaps, interview topics, and application strategy.
offer-compare-skill
A decision assistant for comparing two or more job offers across pay, growth, company strength, team risk, promotion, AI exposure, and lifestyle. Total compensation means the combined value of salary, bonuses, shares, and sign-on payments.
offer-toolkit-skill
A set of job-search tools covering the path from finding a vacancy to accepting an offer. It includes separate tools for job searches, job-description analysis, tailored resumes, behavioral interviews, offer comparisons, and salary negotiation.
bq-skill
A coaching workflow for building a reusable library of truthful stories for behavioral job interviews. It uses structures such as STAR—Situation, Task, Action, Result—and can connect stories to a job description and résumé.
linkedin-job-search-skill
A job-search assistant that uses a sample job description and your resume to find and rank similar public LinkedIn Jobs listings. LinkedIn Jobs is LinkedIn’s section for advertised employment positions.