agentleFS
Sign inStart free
ariaxhanariaxhan/kernel-claude/skills/knowledge-graph/SKILL.md

knowledge-graph

Skill15 starsChanged 4 months agoMITView on GitHub

Build, refresh and query a deterministic code knowledge graph to cut orientation-token cost. Triggers: knowledge graph, graphify, code graph, god nodes, orientation cost, map the codebase, what connects, callers of, blast radius.

23 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use push, deploy and 2 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • OpenAISkill

    push

    Also in 3 projects that use this

  • VercelSkill

    deploy

    Deploy applications to Vercel. Use when the user says "deploy", "deploy to Vercel", "push to production", "deploy my app", or "go live".

    Also in 3 projects that use this

  • varun29ankuSSkill

    orchestrate

    This skill should be used when the user asks to 'orchestrate a task', 'break down work into parallel agents', 'coordinate subtasks', 'run agents in parallel', or mentions 'multi-agent'. Decomposes complex tasks into tracked subtasks, dispatches parallel subagents, and coordinates until completion.

    Also in 3 projects that use this

  • nicolasestremSkill

    brainstorm

    Structured idea generation and solution exploration with systematic analysis.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 19 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If graphify is absent, this skill degrades to a no-op — never a hard failure.

    line 33
  2. Very specific:

    NEVER commit graphify-out/ — it is DERIVED.

    line 42
  3. Distinctive:

    Run it incrementally (changed files only), never full-corpus, never on every commit.

    line 77

“Common” is measured: kept by at least 6 in 10 of the 19 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md
<skill id="knowledge-graph"><purpose> An agent's token bill splits into ORIENTATION (finding where the answer lives — reading files, following imports, grepping) and REASONING (actually solving). On a large, tangled repo the orientation half dominates, and it is pure overhead: the model is not thinking yet, it is still navigating. A pre-built code knowledge graph replaces that file-crawl with one query, so you pay orientation tokens once (at build) instead of every session.

The saving is CONDITIONAL on repo size × tangle, not a fixed multiplier. Measured on real repos: ~5.7x fewer tokens/query on a mid-size service, ~73x on a large interconnected one, ~13% on a tiny library. The graph query cost is ~constant; naive full-corpus cost scales with size — so reduction = corpus ÷ constant. Do the arithmetic on YOUR repo, don't quote a headline.

Hard boundary: the graph helps NAVIGATION, not REASONING. "Design a cache", "why is this slow" get zero lift. It gathers context efficiently; it does not think for the model. </purpose>

<prerequisite> Uses `graphify` (open-source, tree-sitter + NetworkX, MIT). Code extraction is local + deterministic + free (no API key). Install once: `uv tool install graphifyy` (or `pipx`/`pip`). If graphify is absent, this skill degrades to a no-op — never a hard failure. </prerequisite><build> Code-layer graph (free, offline, seconds): ```bash graphify extract <path> --code-only # local AST only; skips docs; no LLM, no cost ``` Outputs `graphify-out/graph.json` (+ report; +interactive graph.html under ~5000 nodes). NEVER commit `graphify-out/` — it is DERIVED. Gitignore it and rebuild on demand (the sqlite-mirror discipline: commit the source, rebuild the artifact). </build><query> ```bash graphify query "what connects auth to the database?" # BFS over the graph, token-budgeted graphify path "UserService" "DatabasePool" # shortest path between two symbols graphify god-nodes --top 12 # architectural hubs (most-connected) graphify affected "RateLimiter" # reverse traversal = change blast radius graphify benchmark # measure YOUR token reduction, per question ``` `god-nodes` doubles as a comprehension + pruning lens: hubs are the real spine; low-degree, never-linked nodes are dead-code / consolidation candidates. Also available as an MCP server (`query_graph`, `shortest_path`, `get_neighbors`) for repeated structured access. </query><automatic> The graph pays off only if it is CONSULTED. A skill telling the agent to reach for it is opt-in and unreliable, so the orientation layer is AMBIENT: when the working repo has a graph, the session-start hook injects its architectural spine (top god-nodes + the query commands) directly into context — the agent boots already oriented, no tool call, no human ask. Deep on-demand queries ("what calls this exact function") still go through `graphify query`/`path`/`affected` or the MCP tools; those are available + steered, but the baseline map arrives for free. </automatic><continuous> A stale graph is worse than none. Keep it fresh, cheaply: - **Code layer (free):** `graphify extract <path> --code-only` is incremental via its AST cache ("N cached/unchanged, 0 re-extracted"). Wire it into `post-commit` so the graph is never more than one commit stale, at ~zero cost. Opt-in installer: `hooks/scripts/knowledge-graph.sh install` (gated on `KERNEL_GRAPH_ON=1`, mirroring autopush — never stamps hooks by surprise). - **NEVER `graphify update`** for the code graph: it re-scans ALL files and adds docs as bare nodes (measured 1506 → 13156 on one tree). Always `extract --code-only`. - **Doc/semantic layer** (summaries, tags, prose edges) needs a model and is OPTIONAL polish. Run it incrementally (changed files only), never full-corpus, never on every commit. Local models are "good enough for orientation, not a top-tier artefact"; a frontier model is sharper. </continuous><boundaries> - Free + deterministic is the CODE layer only. The doc/paper/image "why" layer sends semantic descriptions (never raw source) to a configured backend — that costs a model. - Small or reasoning-heavy repos: the graph is a solved problem you did not have. Skip it. - The graph is a comprehension artefact that also saves tokens — value it as a map first. </boundaries></skill>

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free