Imported from jce529/NovelScript (.claude/skills/graphify/SKILL.md). Install upstream with npx skills add jce529/NovelScript --skill graphify. Copyright stays with the author.

graphify

Name: graphify
Author: jce529

This skill is a pointer. The canonical instructions live in .codex/skills/graphify/, shared with the Codex CLI so both agents follow one copy and stay in sync when graphify is upgraded.

How to use this skill

Read .codex/skills/graphify/SKILL.md in full — that is the real skill body.

Read the reference file that matches the task:

File	Covers
`references/query.md`	`query` / `path` / `explain` — reading the graph
`references/update.md`	`update` — incremental rebuild after code changes
`references/extraction-spec.md`	how nodes and edges are extracted
`references/exports.md`	SVG, GraphML, Neo4j, FalkorDB, wiki, Obsidian
`references/hooks.md`	post-commit auto-rebuild hook
`references/add-watch.md`	`add <url>` and `--watch`
`references/transcribe.md`	audio/video ingestion
`references/github-and-merge.md`	cloning repos and cross-repo merges

All paths are relative to .codex/skills/graphify/.

Environment (verified on this machine)

CLI: graphify 0.9.56, installed as the Python package graphifyy under C:\Users\chang\AppData\Local\Programs\Python\Python313. Already on PATH.
Graph data: graphify-out/ (relative paths only, so it is portable).
Skipping the install-detection block in Step 1 of the canonical SKILL.md is fine while graphify --version succeeds.

Project rules

AGENTS.md (imported by CLAUDE.md) carries the binding rules — prefer graphify query over raw grep for codebase questions, and run graphify update . after modifying code. Those rules win over anything here.

{ "ocm": "1", "id": "jce529-novelscript-graphify", "kind": "skill", "name": "graphify", "description": "Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a persistent knowledge graph with god nodes, community detection, and query/path/explain tools.", "publisher": "jce529", "version": "1.0.0", "capabilities": { "domains": [ "data_analysis", "coding" ], "tags": [ "skill-md", "github" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a persistent knowledge graph with god nodes, community detection, and query/path/explain tools." ], "primary": false, "metadata": { "source": { "provider": "github", "repository": "https://github.com/jce529/NovelScript", "path": ".claude/skills/graphify/SKILL.md", "ref": "4350c466cbda2e2f0f9db9ea1fc0dd297d4424a8", "url": "https://github.com/jce529/NovelScript/blob/4350c466cbda2e2f0f9db9ea1fc0dd297d4424a8/.claude/skills/graphify/SKILL.md", "key": "jce529/NovelScript/.claude/skills/graphify/SKILL.md" } }, "instructions": "# graphify\n\nThis skill is a pointer. The canonical instructions live in `.codex/skills/graphify/`,\nshared with the Codex CLI so both agents follow one copy and stay in sync when\n`graphify` is upgraded.\n\n## How to use this skill\n\n1. Read `.codex/skills/graphify/SKILL.md` in full — that is the real skill body.\n2. Read the reference file that matches the task:\n\n | File | Covers |\n |---|---|\n | `references/query.md` | `query` / `path` / `explain` — reading the graph |\n | `references/update.md` | `update` — incremental rebuild after code changes |\n | `references/extraction-spec.md` | how ", "cost": { "context_tokens": 412 } }

graphify

About

graphify

How to use this skill

Environment (verified on this machine)

Project rules

Use it

Manifest

Reviews