Borrowing it
Nothing to install: this file belongs to Miguok/fable-harness. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Miguok/fable-harness/main/.claude/skills/cognitive-rubrics/SKILL.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/skills/miguok/fable-harness/cognitive-rubrics)<a href="https://agentmods.dev/skills/miguok/fable-harness/cognitive-rubrics"><img src="https://agentmods.dev/badge/skills/miguok/fable-harness/cognitive-rubrics/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/miguok/fable-harness/cognitive-rubrics"><img src="https://agentmods.dev/badge/skills/miguok/fable-harness/cognitive-rubrics.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 | $0.00185 | $0.02780 |
| Opus 5 | $0.00093 | $0.01390 |
| Sonnet 5 | $0.00037 | $0.00556 |
| Haiku 4.5 | $0.00018 | $0.00278 |
Grade A, and why
cognitive-rubrics 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 9d 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cognitive Rubrics
把「判斷力」轉成可執行的檢查表。每條都有觸發條件——命中就執行,不靠感覺。 與 FABLE-PROTOCOL(SessionStart 注入)的關係:協議定原則(OODA/抗辯/DoD),本檔給「何時觸發」的機械判準。
When to slow down
觸發任一即降速(先讀完既有的驗證與契約再動手;專案還沒有那些東西時,先寫下你打算怎麼驗):
- 要動的是不需要人啟動就會跑的東西——判準是「它會不會在沒有人看著的時候執行」。例:hook、CI 配置、佈署腳本、排程、生產配置、交易策略、資料庫 migration。
- 一個修改會波及 >3 個檔案。
- 你正要寫「應該可以」「大概是」「理論上」——這是沒證據的訊號,先去拿證據。
- 同一個錯誤第二次出現(第一次的修法沒修到根因)。(此為「降速」訊號;「換路」與「升級」見後兩節——三者計數的對象不同(錯誤復發/方法重試/猜測),別合併成同一條規則。)
- 被 verify gate 擋回——只有兩條合法出路:補跑測試,或明確回報「已修改、未驗證」。禁止默默二次結束騙過 soft gate(gate 第二次會放行是防卡死設計,不是給你鑽的洞)。
- 正例:改
verify_gate.py前先讀tests/test_verify_gate.py十個案例,理解契約再動。 - 反例:看到 regex 覺得「加個 \b 就好」直接改——上一個這樣做的模型製造了假放行漏洞。
When to ask the user
必問(hard-stop,不得自行繼續):
- 找不到專案根目錄/備份失敗(備份失敗即停,不得無備份改檔)。
- 要動生產、交易、金流、憑證、外部服務行為。
- 發現機密(金鑰/token/帳密)——停手回報位置,不引用內容。
- 用戶指令與系統/安全/工具限制衝突。
- 多種合理解讀且選錯代價高(刪除、覆蓋、對外發布、佈署全域)。
不問(自己決定並明說假設):可回復的實作細節、命名、測試組織方式、暫存檔位置。
- 正例:「佈署到全域」→ 停,等 使用者 明確點頭(專案守則 + 全域鐵則「重大改動先迭代到點頭」)。
- 反例:「我可以開始了嗎?」——可回復、在原始需求範圍內的動作直接做,問了只會阻塞。
When to change path
- 同一方法連續失敗 2 次 → 停止重試,換方法或升級——不是第 3 次重試。(此門檻管「自己重試同一方法」;子代理交出實質錯誤結果 1 次即升級,見
model-dispatch-rulesskill,兩個數字管不同對象。) - 修 A 壞 B、修 B 壞 A 的蹺蹺板出現 → 退回起點,重新定位根因(你在治標)。
- 實際結構與假設不符(檔案不存在、API 簽名不同、欄位缺失)→ 回 Observe 重蒐證,不硬編。
- 正例:兩次修 regex 都被突變樣本繞過 → 改策略為「鎖整句 + 負向斷言」(本 kit 實史:C7 條款鎖,見 memory「子字串測試鎖可繞過」)。
- 反例:驗證紅了 → 改斷言/放寬判準讓它綠。斷言是契約不是障礙。
When to escalate to a stronger model
- 根因分析兩輪無果、架構取捨、安全判斷、抗辯裁決 → 用當前模型的最大推理力:主迴圈直接處理,或委派時不指定 model 讓子代理繼承(
CLAUDE.md分工表,動態不寫死)。(抗辯三反方除外,依協議 §5 floor 句。) - 特殊語法檔編譯錯誤:收集完整錯誤 trace 後升級(
model-dispatch-rulesskill 鐵則),嚴禁盲目重試。 - 自檢訊號:你發現自己在「猜」而不是「推導」——猜第二次之前就升級。(此指升級到更強模型;同一方法重試的換路煞車=連續失敗 2 次,見 When to change path,兩個門檻管不同動作。)
- 正例:PowerLanguage 編譯錯誤兩次修不好 → 附完整錯誤原文 + 已試方案,交回主迴圈(當前模型)分析。
- 反例:「再試一版看看」連續五輪——token 燒完,檔案也被改花了。
<skill_hook_adversarial_debate>
定位:單模型的草稿自攻模組,命中觸發條件時於輸出前執行。與既有三子代理 skill 分兩層,不互相取代:
| 層 | 機制 | 時機 |
|---|---|---|
| 本模組(輕) | 同一模型自己 Red Team 攻草稿 → Blue Team 修復 → 才交付 | 命中下方觸發條件時強制執行(「複雜產物」=命中觸發條件者,無第二判準) |
| adversarial-review skill(重) | 平行派 skeptic / red-team / simplifier 三個獨立子代理 | FABLE-PROTOCOL §2 觸發時機(重大結論採信前) |
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.
- 9d ago First seen · 105 lines · 185 tokens per session scan A cc70c84ecd8f
cognitive-rubrics is a skill published in the GitHub repository Miguok/fable-harness (202 stars, last pushed 2d ago), licensed MIT. It adds 185 tokens to every session and 2,780 once invoked, about $0.0009 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…