New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Graph Skills Retriever

by davidliuk in davidliuk/graph-of-skills

Retrieve a bounded bundle of relevant external skills from a prebuilt Graph of Skills workspace.

MITAuto-check passedKnowledge Management

Install Graph Skills Retriever

skills CLI
$ npx skills add davidliuk/graph-of-skills --skill graph-skills-retriever -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install davidliuk/graph-of-skills graph-skills-retriever --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/davidliuk/graph-of-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/graph-skills-retriever .claude/skills/graph-skills-retriever && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
graph-skills-retriever
GitHub stars
219
Token cost
~1.2k tokens
SKILL.md length
508 words
Files
3 (incl. references)
Skills in repo
3
Repo updated
First seen
Licence
MIT

At a glance

Retrieve a bounded bundle of relevant external skills from a prebuilt Graph of Skills workspace.

  • Works in 8 steps: Start with a short task-level query. → Read the returned bundle carefully and… → If the result says Retrieval Status:… → …
  • The task may need specialized skills
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • References that are not already obvious from current context

What it does

Graph Skills Retriever is an agent skill from davidliuk/graph-of-skills. Retrieve a bounded bundle of relevant external skills from a prebuilt Graph of Skills workspace. Use when the task may need specialized skills, scripts, or references that are not already obvious from current context, especially in containerized eval environments that mount a prebuilt GoS graph.

Its SKILL.md is about 1.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `agents/openai.yaml` and `references/container-layout.md`). Compatibility notes: ["claude-code","codex","gemini-cli"]

It sits in Knowledge Management. The repository describes itself as: [EMNLP '26] Dependency-Aware Structural Retrieval for Massive Agent Skills. The licence is MIT.

When your agent uses it

  • The task may need specialized skills
  • References that are not already obvious from current context
  • Especially in containerized eval environments that mount a prebuilt GoS graph

Example prompts

  • “/graph-skills-retriever”

Requirements

  • Compatibility (from SKILL.md): ["claude-code","codex","gemini-cli"]
  • Pre-approved tools (allowed-tools): shell

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Start with a short task-level query.
  2. Read the returned bundle carefully and check the retrieval status.
  3. If the result says Retrieval Status: NO_SKILL_HIT, explicitly state that no relevant skill was retrieved and continue on a no-skill path…
  4. If the result says Retrieval Status: SKILL_HIT, inspect the task requirements, tests, and verifier first. Write down the minimum…
  5. Use the exact Source: paths returned by retrieval. Do not reconstruct paths from the skill name or scan the whole skill library if a…
  6. Follow the retrieved skill instructions and inspect the referenced Source: paths when you need scripts or references from a specific skill.
  7. Use the skill bundle to narrow the solution space. Prefer the shortest path to verifier pass, and prefer adapting an existing script or…
  8. Re-query with a narrower subproblem if the task shifts.

What it can do on your machine

Read from SKILL.md and the folder at commit 203f60a. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • shell

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are bash).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

  • Compatibility

    ["claude-code","codex","gemini-cli"]

    From compatibility in the SKILL.md frontmatter.

Context cost

Graph Skills Retriever loads about 1.2k tokens when it runs, and up to ~1.6k if it reads all its reference files. Until then it costs about 80 tokens; SKILL.md has 508 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~80
When it runs · the whole SKILL.md, loaded when a task matches
~1.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from davidliuk/graph-of-skills at commit 203f60a, republished under its MIT licence (© davidliuk). 508 words, ~1,191 tokens.

Download SKILL.mdSave it as .claude/skills/graph-skills-retriever/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
graph-skills-retriever
description
Retrieve a bounded bundle of relevant external skills from a prebuilt Graph of Skills workspace. Use when the task may need specialized skills, scripts, or references that are not already obvious from current context, especially in containerized eval environments that mount a prebuilt GoS graph.
allowed-tools
shell
compatibility
["claude-code","codex","gemini-cli"]
inputs
task description or subproblem summary
outputs
a ranked skill bundle with full instructions and source paths

Purpose

Use this skill instead of manually browsing a large skill library.

It assumes the environment already provides:

  • the graphskills-query CLI
  • a GoS workspace or a prebuilt workspace configured through GOS_WORKING_DIR and optionally GOS_PREBUILT_WORKING_DIR

If that wiring is missing, read references/container-layout.md.

Retrieve Relevant Skills

Construct the query yourself. Do not rely on the retrieval system to infer missing task structure for you.

A good query should usually include only the retrieval-critical fields that are actually known:

  • the concrete goal
  • the main artifact or file format
  • the key operation or algorithm
  • the required library, API, protocol, or tool name if known
  • the verifier-critical constraint or invariant
  • the task object being edited, parsed, generated, optimized, or validated

Keep it short, but make it specific. Prefer a compact noun/verb phrase over a long paragraph.

Good patterns:

text
update embedded xlsx in pptx and preserve formulas
parallel tfidf indexing with processpoolexecutor deterministic ranking
civ6 district adjacency exact calculator for verifier
parse branching dialogue script into graph export

Bad patterns:

text
please solve this task for me
I need help with a benchmark task
fix the project and make everything work

Run:

bash
graphskills-query "short specific query with goal + artifact + operation + constraint"

Useful flags:

bash
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
graphskills-query "review paper references and improve the draft" --workspace /opt/graphskills/runtime

How To Use The Results

  1. Start with a short task-level query.
  2. Read the returned bundle carefully and check the retrieval status.
  3. If the result says Retrieval Status: NO_SKILL_HIT, explicitly state that no relevant skill was retrieved and continue on a no-skill path. Do not imply that you used a retrieved skill.
  4. If the result says Retrieval Status: SKILL_HIT, inspect the task requirements, tests, and verifier first. Write down the minimum acceptance requirements before implementing.
  5. Use the exact Source: paths returned by retrieval. Do not reconstruct paths from the skill name or scan the whole skill library if a Source: path is already available.
  6. Follow the retrieved skill instructions and inspect the referenced Source: paths when you need scripts or references from a specific skill.
  7. Use the skill bundle to narrow the solution space. Prefer the shortest path to verifier pass, and prefer adapting an existing script or interface over inventing a broader replacement.
  8. Re-query with a narrower subproblem if the task shifts.
Show full SKILL.md (200 more words)Show less

Guidance

  • Prefer 1-2 targeted retrieval calls over scanning the whole library.
  • Keep the query focused on the current task or subproblem, not the whole conversation history.
  • Query content priority: goal > artifact/format > operation/API > verifier constraint.
  • Include filenames, formats, protocols, or library names when they are part of the task signal.
  • Include exact invariants when they matter, e.g. preserve formulas, deterministic ranking, exact total, match verifier.
  • Do not include benchmark names, generic filler, or conversation meta-text unless they are truly task-relevant.
  • If the result is too broad, narrow the query and reduce --top-n.
  • If the result is empty, retry with simpler keywords before giving up.
  • If no skill is retrieved after retrying, say so explicitly and solve without pretending a skill was used.
  • After a skill hit, take the shortest path to verifier pass and satisfy only the verifier's minimum requirement first.
  • Use the exact Source: paths already returned before searching elsewhere.
  • Do not add extra features, side outputs, UI panels, or refactors unless the task explicitly requires them.
  • Treat retrieved skills as a constraint on implementation choices, not permission to explore more branches.
  • Use --json only when you need structured fields like scores or edge evidence; plain text is usually enough.

© davidliuk, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files (references) in skills/graph-skills-retriever of davidliuk/graph-of-skills.

  • SKILL.md
  • agents/openai.yaml
  • references/container-layout.md

Open the folder on GitHubat commit 203f60a

Compare with similar skills

Graph Skills Retriever next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Graph Skills Retriever compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Graph Skills Retriever this skilldavidliuk/graph-of-skills219—~1.2kAutomated safety check: PassMIT
Logseq Review Workflow Evallogseq/logseq45k—~1kAutomated safety check: PassAGPL-3.0
Baoyu URL To Markdownsdyckjq-lab/llm-wiki-skill2.5k2 repos~3.2kAutomated safety check: PassNone
Obsidian CLIAtmosphere/atmosphere3.8k13 repos~795Automated safety check: PassApache-2.0
Esm Cjs Risk Scanlogseq/logseq45k—~3.3kAutomated safety check: PassAGPL-3.0
Knowledge Searchdataelement/bisheng12k—~1.1kAutomated safety check: PassApache-2.0

Similar skills

  • Logseq Review Workflow Eval

    logseq/logseq

    Compare two revisions of the Logseq logseq-review-workflow skill by running the same review prompt against isolated before and after skill snapshots, collecting both outputs, and producing a…

    45k GitHub stars~1k tokensUpdated today
    Knowledge ManagementAuto-check passed

  • Baoyu URL To Markdown

    sdyckjq-lab/llm-wiki-skill

    Fetch any URL and convert to markdown using Chrome CDP. An agent skill from sdyckjq-lab/llm-wiki-skill.

    2.5k GitHub starsUsed in 2 repos~3.2k tokens
    Knowledge ManagementAuto-check passed

  • Obsidian CLI

    Atmosphere/atmosphere

    Interact with Obsidian vaults using the Obsidian CLI to read, create, search, and manage notes, tasks, properties, and more.

    3.8k GitHub starsUsed in 13 repos~795 tokens
    Knowledge ManagementAuto-check passed

  • Esm Cjs Risk Scan

    logseq/logseq

    Scan Logseq ClojureScript Node/Electron targets for npm module loading risks, especially ESM-only packages that may fail when loaded through js/require or shadow-cljs require-based shims.

    45k GitHub stars~3.3k tokensUpdated today
    Knowledge ManagementAuto-check passed

  • Knowledge Search

    dataelement/bisheng

    Search the user's knowledge bases and knowledge spaces (企业知识库检索).

    12k GitHub stars~1.1k tokensUpdated yesterday
    Knowledge ManagementAuto-check passed

  • Capture Conversation

    outline/outline

    Save the current conversation, a decision, or a set of notes as a document in an Outline collection; use when the user wants to keep what was discussed in their knowledge base.

    41k GitHub stars~474 tokensUpdated today
    Knowledge ManagementAuto-check passed

More from davidliuk/graph-of-skills

  • Graph Skills Retriever

    davidliuk/graph-of-skills

    Retrieve a bounded bundle of relevant external skills from the local Graph Skills library instead of manually browsing hundreds of skills.

    219 GitHub stars~437 tokensUpdated 1 mo ago
    Auto-check passed

  • Vector Skills Retriever

    davidliuk/graph-of-skills

    Retrieve a bounded bundle of relevant external skills from the local skill workspace using vector embedding similarity only.

    219 GitHub stars~440 tokensUpdated 1 mo ago
    Auto-check passed

Categories

Questions about Graph Skills Retriever

What does Graph Skills Retriever do?

Retrieve a bounded bundle of relevant external skills from a prebuilt Graph of Skills workspace. Graph Skills Retriever is an agent skill from davidliuk/graph-of-skills. Retrieve a bounded bundle of relevant external skills from a prebuilt Graph of Skills workspace.

When should I use Graph Skills Retriever?

Graph Skills Retriever fits situations like: the task may need specialized skills; references that are not already obvious from current context; especially in containerized eval environments that mount a prebuilt GoS graph.

How do I install Graph Skills Retriever in Claude Code?

Run `npx skills add davidliuk/graph-of-skills --skill graph-skills-retriever -a claude-code`. Or copy the skill folder (skills/graph-skills-retriever in davidliuk/graph-of-skills) into .claude/skills/graph-skills-retriever in your project. Claude Code loads it when a task matches its description.

How do I install Graph Skills Retriever in Codex?

Run `npx skills add davidliuk/graph-of-skills --skill graph-skills-retriever -a codex`. Or copy the skill folder (skills/graph-skills-retriever in davidliuk/graph-of-skills) into .agents/skills/graph-skills-retriever in your project. Codex loads it when a task matches its description.

Can I use Graph Skills Retriever in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add davidliuk/graph-of-skills --skill graph-skills-retriever -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/graph-skills-retriever, .gemini/skills/graph-skills-retriever, .github/skills/graph-skills-retriever and .opencode/skills/graph-skills-retriever in your project.

What does Graph Skills Retriever need to run?

SKILL.md names no scripts, command-line tools or credentials: Graph Skills Retriever is instructions for the agent only. Its frontmatter pre-approves these tools: shell. Compatibility (from SKILL.md): ["claude-code","codex","gemini-cli"].

Does Graph Skills Retriever access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Graph Skills Retriever safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Graph Skills Retriever use?

Graph Skills Retriever is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Graph Skills Retriever use?

About 1.2k tokens (SKILL.md is roughly 4.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 366 tokens, read only when the agent opens those files.

What are the alternatives to Graph Skills Retriever?

Skills that share tags, products or a category with Graph Skills Retriever: Logseq Review Workflow Eval (logseq/logseq, 45k stars), Baoyu URL To Markdown (sdyckjq-lab/llm-wiki-skill, 2.5k stars), Obsidian CLI (Atmosphere/atmosphere, 3.8k stars) and Esm Cjs Risk Scan (logseq/logseq, 45k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Graph Skills Retriever?

davidliuk (a GitHub user) maintains it in davidliuk/graph-of-skills, which has 219 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on August 30, 2026.

Source: davidliuk/graph-of-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.