界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
开发与工程 · SKILL.md

graph-skills-retriever

graph-skills-retriever

【开发与工程】Retrieve a bounded bundle of relevant external skills from the local Graph Skills library instead of manually browsing hundreds of skills.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: graph-skills-retriever description: Retrieve a bounded bundle of relevant external skills from the local Graph Skills library instead of manually browsing hundreds of skills. inputs:

  • task description or subproblem summary outputs:
  • a ranked skill bundle with full instructions and source paths under /opt/graphskills/library compatibility:
  • claude-code
  • codex
  • gemini-cli allowed-tools:
  • shell
  • python3

Purpose

Use this skill when the task seems to require specialized domain knowledge, scripts, or references that are not already obvious from the current context.

The full skill library is stored outside the auto-loaded harness context at:

/opt/graphskills/library

Do not manually scan that whole directory first. Retrieve a focused bundle.

Retrieve Relevant Skills

Run:

graphskills-query "short description of the task or current subproblem"

Useful flags:

graphskills-query "debug spring boot jakarta migration build errors" --top-n 5 --seed-top-k 4 --max-context-chars 9000
graphskills-query "extract text from receipts into xlsx" --json

How To Use The Results

  1. Start with a short task-level query.
  2. Read the returned skill bundle.
  3. Follow the retrieved skill instructions and inspect any referenced files in /opt/graphskills/library/<skill-name>/.
  4. Re-query with a narrower subproblem if needed.

Guidance

  • Prefer 1-2 retrieval calls over manually searching the entire library.
  • The returned Source: path points to the canonical SKILL.md in /opt/graphskills/library.
  • If you need scripts or references from a retrieved skill, inspect that skill directory directly.
  • Reuse the graph retriever whenever the task shifts to a new subproblem.
我的收藏