- name
SKILL.md
技能檔案保留在發布者的儲存庫中。
在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add dongshuyan/compass-skills --skill run-history-skill-builder -a cursor -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add dongshuyan/compass-skills --skill run-history-skill-builder -a claude-code -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add dongshuyan/compass-skills --skill run-history-skill-builder -a codex -y在終端機中貼上複製的指令。它會將這個 SKILL.md 安裝到對應的代理中。
npx skills add dongshuyan/compass-skills --skill run-history-skill-builder -a grok -y在 Grok Bot 聊天中傳送複製的訊息。Bot 會從 GitHub 取得這個 SKILL.md,並加入你所有 Bot 共用的技能中。
Install this skill: https://github.com/dongshuyan/compass-skills/tree/HEAD/skills/run-history-skill-builder來自發布者
Write all user-facing output in the user's language. Default to Chinese when the language is unknown.
Turn real run history into a new skill package, a plan-only skill design, or an upgrade handoff when the request is actually about an existing skill.
This skill is agent-agnostic. It should work in Codex, Claude Code, OpenCode, OpenClaw, Hermes, and similar local agent hosts that can read SKILL.md plus optional references/, scripts/, evals/, and agents/.
<skill-dir> from the directory that contains this SKILL.md.<python> mean the host's available Python launcher: python3, python, or py -3.<skill-dir>/scripts/... describe path segments, not a required separator style. On Windows, use the separator style that your shell or harness accepts.skills/ directory or the host agent's documented local skills directory, then wait for confirmation.plan_only, new_single_skill, router_skill, skill_suite, or existing_skill_upgrade_handoff.Do not jump directly from "I saw a successful run" to "I wrote a skill". The missing middle layer is where portability, privacy, and generalization are decided.
plan_only: the user wants a reviewed design or audit, not files.new_single_skill: one stable workflow or one tightly coupled workflow family.router_skill: one entry point that routes across several existing skills or phases.skill_suite: several independent workflows that should be released together but triggered separately.existing_skill_upgrade_handoff: the real task is to improve an existing skill. Produce a clean handoff for $run-history-skill-upgrader instead of editing that skill here.Prefer replacement, merging, and omission over package bloat.
Allowed by default after intent is locked:
Require explicit approval before reading:
Keep facts, inferences, and open assumptions separate. Never write secrets, hidden prompts, private account identifiers, or unrelated personal data into the released skill or its examples.
SKILL.md focused on trigger boundary, role, workflow, safety gates, and reference navigation.references/.scripts/.evals/ when the workflow is long-lived, high-risk, or easy to overfit.README, installation scripts, changelogs, or decorative files unless the user or release target explicitly requires them.agents/openai.yaml as an optional UI enhancement, not as the core logic.Run the package validator bundled with this skill:
<python> <skill-dir>/scripts/validate_skill_package.py <target-skill-dir>
This bundled validator checks package structure, the portable Agent Skills frontmatter field set, referenced paths, JSON shape, Python syntax, and common private-path leaks. Its dependency-free frontmatter preflight accepts scalar fields, block text, and one-level string metadata; it rejects other YAML forms instead of guessing. A specific host may accept different syntax or a narrower field set, so its canonical validator remains authoritative for installation there. Neither structural check runs the eval cases or proves that the skill triggers correctly or improves behavior.
If the current host provides a canonical skill validator, run that too. On Codex-like hosts, this often means a quick_validate.py command from the platform's skill tooling.
Also run the smallest relevant technical checks:
python -m py_compile for modified Python scripts;python -m json.tool for edited JSON files;Do not claim completion if validation was skipped or failed. Report the gap and the remaining risk.
Report:
plan_only, a new skill package, or an upgrader handoff.references/history-mining.mdreferences/open-source-pattern-mining.mdreferences/skill-design-protocol.mdreferences/self-repair-and-evals.mdreferences/examples.mdscripts/validate_skill_package.pyevals/evals.jsonTurn a completed task, browser flow, artifact pipeline, failure-recovery trace, or repeatedly refined workflow into a new reusable skill package or a reviewed skill-design plan. Use when the user asks
---name: run-history-skill-builderdescription: Turn a completed task, browser flow, artifact pipeline, failure-recovery trace, or repeatedly refined workflow…---# skills/run-history-skill-builder/SKILL.md總覽
Turn a completed task, browser flow, artifact pipeline, failure-recovery trace, or repeatedly refined workflow into a new reusable skill package or a reviewed skill-design plan. Use when the user asks to make a new skill from real run history, extract a reusable workflow from conversation/logs/files, summarize lessons into a new skill, or produce a plan before writing files. Do not use to upgrade an existing skill or to execute the business workflow itself.
常見問題
從發布者下載 skills/run-history-skill-builder/SKILL.md,或複製路徑後開啟儲存庫。PlugMCP 只收錄這項技能,不代管檔案。
發布者將它存放在 https://github.com/dongshuyan/compass-skills。
Claude、Cursor、Codex 和 Grok Build 可以讀取你從發布者儲存庫加入的 SKILL.md。在聊天中傳送技能的 GitHub 資料夾連結,Grok Bot 就會安裝它。