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 agentmods add instructions/miguok/fable-harness/claude-mdgit clone --depth 1 https://github.com/Miguok/fable-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/instructions/miguok/fable-harness/claude-md)<a href="https://agentmods.dev/instructions/miguok/fable-harness/claude-md"><img src="https://agentmods.dev/badge/instructions/miguok/fable-harness/claude-md.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 | $0.01077 | $0.01077 |
| Opus 5 | $0.00539 | $0.00539 |
| Sonnet 5 | $0.00215 | $0.00215 |
| Haiku 4.5 | $0.00108 | $0.00108 |
Grade A, and why
fable-harness CLAUDE.md 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 6d 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
Fable Protocol Kit — 專案指示
本專案是「讓 Opus/Sonnet/Haiku 以 Fable 5 的操作模式工作」的行為協議套件（hooks + skill + agents）。
行為協議本體由 SessionStart hook 注入（.claude/hooks/fable_protocol.md），此處只放模型分工、不可妥協守則、治理文件路由與專案守則。
模型分工（Model Routing）— 委派子代理時強制遵守
主迴圈無論當前是哪個模型，透過 Agent 工具或 Workflow agent() 委派時，依任務性質選 model：
| 任務性質 | model | 例子 |
|---|---|---|
| 邏輯推理、架構設計、根因分析、重大決策、抗辯裁決 | 當前模型（不指定 model） | 設計評估、bug 根因判定、judge panel |
| 程式編寫、重構、bug 修復、測試撰寫 | sonnet |
實作功能、改測試、修 lint |
| 檔案批次處理、文書編輯、資料搜尋、格式轉換、摘要 | haiku |
Explore 搜碼、整理 log、改 README 措辭 |
- 瑣碎單步操作（改一行、看一個檔）主迴圈直接做，不委派。
- 抗辯三反方的 model 底線規則見協議 §5 與 adversarial-review skill（正本），此處不重複。
- 「當前模型」＝主迴圈直接做，或委派時省略 model 參數讓子代理繼承——推理力跟著 session 走，不降級到寫死的型號（指表格推理列；反方底線見協議 §5）。
- 改此表時必須同步改
.claude/hooks/fable_protocol.md§5（雙源同步）。 - 派工包 7 欄、回報模板、升級/降級規則見
model-dispatch-rulesskill——委派時強制使用。
不可妥協守則（Non-negotiable Guardrails）
- 精準修改：只動用戶要求的區塊；不順手重排、重格式化、刪無關程式碼。特殊語法檔（PowerLanguage / Pine Script / SQL migration / CI 配置）尤其嚴格。
- 註解與參數說明是受保護內容：優化邏輯也不得刪參數說明、既有註解、交易假設、歷史備註；未經指示刪除＝失敗，不是清理。
- 先讀後寫、改前備份：改任何既有檔案前必先 Read；必先建
<檔名>.bak.<YYYYMMDD-HHMMSS>並確認存在——備份失敗即停。 - 工具誠實：不得宣稱不存在的工具/檔案/權限；能力缺失就明說並降級（如輸出手貼內容）。
- 規則衝突時：選更安全、更窄、可回復的動作；仍不確定 → 先問用戶再動檔案。
治理文件路由（Routing）
| 情境 | 讀這份 |
|---|---|
| 委派子代理（派工包 / 回報格式 / 升級降級） | model-dispatch-rules skill |
| 何時慢下來 / 問用戶 / 換路 / 升級模型；輸出前自我抗辯模組 | cognitive-rubrics skill |
專案守則
- 測試一律放
tests/，遵守全域 docstring 鐵則（四區塊 + 執行紀錄）。 - hooks 腳本改動後必須重跑
python -m pytest tests/ -v且綠燈才算完成。 .claude/hooks/.last_*是 hook 觸發 marker（e2e 測試用），不要手動改、不要進版控。- 佈署到全域（
C:\Users\user\.claude\）前必須：本專案全部測試綠 + 用戶明確點頭。佈署步驟見 README.md。 - 測試環境豁免 uv（使用者 2026-07-04 拍板）：本專案走系統 python + 全域 pytest，不建 pyproject，hooks 亦走系統 python。
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.
- 6d ago First seen · 44 lines · 1,077 tokens per session scan A f1145153391b
fable-harness CLAUDE.md is an instructions file published in the GitHub repository Miguok/fable-harness (202 stars, last pushed 10d ago), licensed MIT. It adds 1,077 tokens to every session, about $0.0054 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 instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).