model-dispatch-rules

model-dispatch-rules is a skill for Claude Code, Codex from Miguok/fable-harness. It costs 149 tokens per session (2,032 once invoked), scanned A, original, MIT.

Rules for assigning coding and exploration work to helper agents. They define what the assignment must contain, when to delegate, which model level to use, and how results should be reported.

In plain words
What is it for?
Use them when searching many files, changing code or tests, running parallel work, creating a worktree, or checking whether a helper agent stayed within scope.
Why use it?
They prevent the main agent from doing large repository scans or poorly defined parallel work. Clear scope, allowed files, checks, and stop conditions make delegated work easier to review.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/miguok/fable-harness/model-dispatch-rules
Any agent
npx skills add Miguok/fable-harness --skill model-dispatch-rules
Clone the repo
git clone --depth 1 https://github.com/Miguok/fable-harness

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for model-dispatch-rules

README.md
[![agentmods](https://agentmods.dev/badge/skills/miguok/fable-harness/model-dispatch-rules.svg)](https://agentmods.dev/skills/miguok/fable-harness/model-dispatch-rules)
Your own site
<a href="https://agentmods.dev/skills/miguok/fable-harness/model-dispatch-rules"><img src="https://agentmods.dev/badge/skills/miguok/fable-harness/model-dispatch-rules.svg" alt="Measured on agentmods" height="20"></a>
Per session 149 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,032 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00149 $0.02032
Opus 5 $0.00075 $0.01016
Sonnet 5 $0.00030 $0.00406
Haiku 4.5 $0.00015 $0.00203

Measured 4d ago against content hash a84bcad7d2a5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

model-dispatch-rules 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 4d 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.

.claude/skills/model-dispatch-rules/SKILL.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Model Dispatch Rules

一句話:主迴圈是指揮官——負責分析、派工、裁決;粗活與編碼派子代理,主迴圈只收結論、風險、決策。 模型選擇以 CLAUDE.md 分工表為準(唯一正本,此處不複製表內容以免雙源漂移;反方底線例外,正本見協議 §5 floor 句);本檔規定「怎麼派」。

使用者 指示(2026-07-04):非瑣碎的程式編寫與測試撰寫一律派 sonnet 執行;主迴圈(無論當前模型多強)負責分析與 workflow 派工。瑣碎單步(改一行、看一個檔)依下方例外由主迴圈直接做——「一律」不含瑣碎單步,兩條規則不衝突。

Commander does not do bulk work

  • 主對話在 Agent / Explore / Workflow 可用時,不得自己做大範圍 repo 掃描、批次讀檔、整目錄 grep。
  • 觸發委派的機械判準(任一命中即派):
    • 預估要 Read >10 個檔
    • Grep 結果 >100 行還需要逐檔深入
    • 任務形如「找出所有 X」「盤點整個 Y」→ 派 Explore(haiku)
    • 要寫或改程式碼/測試 → 派 sonnet(附派工包)
  • 主對話收到的應是:結論 + 風險 + 建議決策 + 關鍵 file:line——不是原始檔傾倒。
  • 例外:≤3 個已知路徑的精準讀取、改一行看一檔的瑣碎單步,主迴圈直接做(委派 overhead 反而更貴)。
  • 正例:「找出所有呼叫 deploy() 的地方」→ 派 Explore,收回 file:line 清單 + 一句摘要。
  • 反例:主迴圈自己 Glob **/* 然後 Read 30 個檔「以防萬一」——context 被灌爆,判斷品質下降。

Dispatch packet

每次委派必含 7 欄,缺任一欄=無效派工,子代理應先要求補齊而非開工:

目標 Goal:<一句話,可驗證,不是「幫我看看」>
範圍 Scope:<這次任務動什麼;一句話界定>
非目標 Non-goals:<明確不要做的事,至少 1 條(防順手改)>
允許路徑 Allowed paths:<可讀清單 + 可寫清單;寫入預設禁止,除非明列>
驗收標準 Acceptance criteria:<可機械判定:一條可重跑的指令+預期輸出。有測試就用測試(「pytest tests/x.py 全綠」);沒有就指定別的可觀察結果(「跑 X 後 Y 檔出現 Z 欄」)。不接受「能動就好」>
回報格式 Report format:<指定下方模板>
停止條件 Stop conditions:<遇到什麼立刻停:範圍外檔案、編譯錯誤、機密、需要刪東西>

唯讀搜尋類輕量派工(如 Explore):Non-goals / Allowed paths / Report format 三欄可填「預設」 (預設=不寫入任何檔案/全 repo 唯讀/本檔回報模板),但欄位本身不可缺。

Required report format

子代理回報一律用此模板(主迴圈收到缺段的回報,退回要求補齊):

結果 TLDR:<一句話>
做了什麼:<動過的檔案清單,每檔一句>
證據:<可重跑的驗證指令 + 原始輸出/量測值;沒有證據就寫「已修改、未驗證」,不得寫「應該可以」>
超出範圍發現:<看到但沒動的問題;只回報不修>
風險與未完成:<明列;真的沒有就寫「無,理由:…」,不得空白>

Escalation and downgrade

  • 一次實質錯誤即升級(此門檻管「子代理交出的結果」;自己重試同一方法的門檻是 2 次,見 cognitive-rubrics skill):子代理(sonnet/haiku)改錯檔、或驗收未過卻回報完成 → 該任務交回主迴圈(當前模型)接手分析,不給第二次盲試。
  • 出界即停:子代理動了 Scope 外的檔案 → 立即停止 → 先保全現狀(複製該檔到 .bak——壞狀態也是證據)→ 定位越界變更、只還原越界的那部分 → 回報「動了 X,已還原」。專案有版控時用 git diff 定位;沒有就靠上一步那份 .bak 比對。警告:整檔還原(git checkout <file>、或整份蓋回備份)會把範圍內的未提交修改一起丟掉且不可復原,禁止當預設動作。
  • 特殊語法檔鐵則(PowerLanguage / EasyLanguage / Pine Script / SQL migration / CI YAML):遇編譯或語法錯誤 → 收集完整錯誤訊息原文 → 升級回報,嚴禁盲目重試語法修補——每次盲試都在燒 token,且常越改越壞、動到不該動的區塊。
  • 降級:推理級任務中發現剩下的是機械性批次動作(改措辭、跑格式轉換)→ 拆出來派 haiku。

Read the full file on GitHub · 89 lines

Changes

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.

  1. 4d ago First seen · 89 lines · 149 tokens per session scan A a84bcad7d2a5

Subscribe to this mod's changes

model-dispatch-rules is a skill published in the GitHub repository Miguok/fable-harness (201 stars, last pushed 8d ago), licensed MIT. It adds 149 tokens to every session and 2,032 once invoked, about $0.0007 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens