TypingMindTypingMindThe best frontend for AI
Run Benchmark Driven Rd logo

Run Benchmark Driven Rd

Community
Hao0321
run-benchmark-driven-rd

可否證、benchmark-gated R&D；可分發專案／Skill 預設涵蓋 GitHub 安全更新、回滾、舊版淘汰。

Overview

PublisherHao0321
Repositoryclaude-skill-code-cleanup
Skill namerun-benchmark-driven-rd
Stars
85
Forks
13
Bundled files
116
LicenseMIT
Links
Skill sourceRepository
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 116 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Hao0321 on GitHub. Read the source before you install it.

Installation

Install the Run Benchmark Driven Rd AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Hao0321/claude-skill-code-cleanup/tree/main/run-benchmark-driven-rd
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Hao0321/claude-skill-code-cleanup.git /tmp/claude-skill-code-cleanup
mkdir -p .claude/skills
cp -r /tmp/claude-skill-code-cleanup/run-benchmark-driven-rd .claude/skills/run-benchmark-driven-rd
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Run Benchmark Driven Rd in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Run Benchmark Driven Rd on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Run Benchmark Driven Rd is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Run Benchmark-Driven R&D

Contract

  1. Active private Skill 唯一權威。先 invocation_revision_gate.py capture 凍結雙 revision，決策前 verify；UNSTABLE 不混用，STALE 重讀重驗。
  2. 修改不越原始授權；本機實作不暗含外部／破壞性 mutation。外部變更須當輪授權、精確 target、preflight、post-readback；GitHub URL 不授權執行。
  3. 先定 claim、baseline、threshold、decision rule，以 task-shaped 正負控制校準 evaluator。工具綠燈不等於品質、完成或 parity；缺同 provenance 證據保持 unmeasured／NOT_CHECKED。
  4. Cleanup 是獨立唯讀 provider；保留 PASS／FAIL／REVIEW／NOT_CHECKED、child、revision evidence；strict promotion 不可改寫缺口為 PASS。
  5. Route receipt 列 intent、stage、artifact/risk、selected/omitted topics、hash、critical IDs、cost、fallback；未知／歧義／stale／超 budget 時 block 或 legacy fallback。
  6. Private learning 存目標 .rd/，不預載；shared rule 需 typed receipt、單一 owner、privacy、正負 fixture、reciprocal link。縮短 context 不刪記憶。
  7. 可分發目標的已授權 implementation／promotion／completion 預設需安全更新能力；audit／security-assessment／source-only 不推導 updater、publish、sign、persistence、delete 權限。
  8. Security route 每個 target 必須規劃並驗完固定六控制；public／parity 再綁同專案 Cleanup、delivery、build 與 release artifact，sibling decoy 無效。
  9. 收尾、重試或交接依 references/disk-hygiene.md 清理授權內可重建舊產物。

路由

任務讀取
Experimentreferences/topics/core-experiment.md
架構／Cleanup baselinereferences/topics/architecture-and-evaluator.md
逆向分析／交付物暴露references/topics/reverse-engineering.md
PR 採用references/topics/community-pr-adoption.md
外部 Skill／pluginreferences/topics/agent-workflow-adoption.md
context／Token／learningreferences/topics/context-and-learning.md
交易／帳號／備份／持久化references/topics/state-integrity.md
3D 角色／UV／Unrealreferences/topics/character-3d-workflow.md
視覺／UI／材質references/topics/visual-quality.md
capability／complete／externalreferences/topics/completion-and-external.md
updater／可分發交付references/topics/secure-self-update.md；依風險加 delivery/security，audit／source-only 除外

大 contract 僅 legacy fallback；維護：references/maintenance.md。

工作流

  1. 解析絕對 repo、授權、contract、obligations、failures、.rd/；route 含 typed updateObligation／capability floor。
  2. 以 Cleanup＋native gate 凍結 baseline 與 claim matrix／provenance。
  3. 做最小決定性實驗，保存 executable、stdout/stderr、exit、elapsed、input/environment hash、artifact、failure type。
  4. 原子寫 typed record 到 .rd/；promotion gate 後才更新 shared topic/index。
  5. 重跑相同 benchmark、負控制、delivered journey、obligations；最後修改後重驗 freshness/revision。
  6. 分開報告 measured stronger／likely but unmeasured／unchanged／regressed/open；PASS 數不等於整體變強。
powershell
python scripts/project_profile_gate.py --project <target> --contract <task-contract.json> --output <route.json> --quiet

Skill 下次 invocation／restart 切換；背景、自啟、破壞性 migration 分開 opt-in。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

and 56 more files.

Frequently asked questions

What does the Run Benchmark Driven Rd AI skill do?

可否證、benchmark-gated R&D；可分發專案／Skill 預設涵蓋 GitHub 安全更新、回滾、舊版淘汰。

Why use Run Benchmark Driven Rd on TypingMind?

Because you install it once and use it with any model. Run Benchmark Driven Rd is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Run Benchmark Driven Rd in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Hao0321/claude-skill-code-cleanup/tree/main/run-benchmark-driven-rd. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Run Benchmark Driven Rd?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Run Benchmark Driven Rd?

As many as you like. As long as a model supports skills, you can use Run Benchmark Driven Rd with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Run Benchmark Driven Rd AI skill free?

Yes. It is published on GitHub by Hao0321 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

Related AI Skills

View all
ruvnet logo

Agent Agentic Payments

ruvnet
CommunityPopular

Agent skill for agentic-payments - invoke with $agent-agentic-payments

8.7K
73.6K
Instructions
View
ruvnet logo

Agent Payments

ruvnet
CommunityPopular

Agent skill for payments - invoke with $agent-payments

8.7K
73.6K
Instructions
View
heygen-com logo

Canopy Part Title

heygen-com
OrganizationPopular

Leaves sweep through the frame and part to reveal the headline. HyperFrames block, 1920×1080, 12s, 11 variables.

6.6K
73.3K
4 files
View
affaan-m logo

Module Regression

affaan-m
CommunityPopular

Govern cross-module regression in large repositories with a REGRESSION.md ledger that records each module's downstream consumers and executable acceptance command. After a change, run the changed…

41.4K
277.3K
2 files
View

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇