skills.rest
laionazeredo
laionazeredoCOMMUNITY
Explore all skills by this author →
GitHub

che-graph

Builds and queries a tree-sitter AST knowledge graph of a codebase via Graphify CLI with fallback engines.

Updated May 12, 2026
Search Tags:#code analysis#knowledge graph#tree-sitter#dependency graph#ast#codebase indexing#graphify
One-click install
npx skills add https://github.com/laionazeredo/che-ai --skill che-graph-laionazeredo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: che-graph
Source: https://github.com/laionazeredo/che-ai/tree/main/skills/che-graph
Command: npx skills add https://github.com/laionazeredo/che-ai --skill che-graph-laionazeredo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires graphifyy.

What problem does it solve? Answering structural questions about a large codebase (where symbols live, how modules depend on each other, which files are import hubs) normally requires slow manual searching; this Skill builds a local AST knowledge graph so agents can query code structure directly. ## Core Features & Use Cases - Knowledge graph refresh: Indexes the repository with Graphify (33 languages, tree-sitter AST) or one of five fallback engines, storing all cache outside the worktree. - Natural language and path queries: Ask questions like "where are auth middlewares defined" or trace a dependency path between two symbols, returning canonical file paths with line ranges. - Stats snapshot: Reports scanned files, language breakdown, and top import hubs for onboarding and code review workflows. - Use Case: Before writing a spec for a refund feature, run a query to find existing refund validation logic and the call path from the router to the Stripe API, avoiding duplicate implementation. ## Quick Start Ask the agent to run /che-graph refresh on this repository and then query where the authentication middleware is defined.

Frequently Asked Questions about che-graph

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a knowledge graph of my codebase?▼

Run the refresh subcommand, which indexes the repository with Graphify CLI (install via pipx install graphifyy) using tree-sitter AST parsing across 33 languages. All cache output is stored outside the worktree in the Che workspace graph directory.

How do I find the dependency path between two functions or files?▼

Use the path subcommand with two symbol names, such as a router handler and an external API call. It returns the ordered call and import chain with file paths and line numbers, falling back to grep-based BFS import analysis if the graph engine lacks path support.

What alternatives to Graphify exist for code knowledge graphs?▼

The fallback chain tries codegraph (21 languages, SQLite FTS5, incremental watcher), @sentropic/graphify, codebase-vis, and @lubab/madar before the Python graphifyy default. If none is installed, a grep-based fallback provides lower-precision structural summaries.

Does the graph cache pollute my git repository?▼

No. All graph output is written to a workspace-level directory outside the worktree, enforced by che assert_outside_worktree checks before any write. The skill never edits .gitignore and never creates artifacts inside the repository.

What happens if no graph engine is installed?▼

The skill prints install options and proceeds with a grep-based fallback that summarizes folder structure and file extension distribution. It never auto-installs engines; the user must manually run one install command.

Related Skills

Explore More Software Engineering Skills→
CloudSenCloudSen

eliteforge-qingtui-cli

Manage QingTui users and send texts via the qingtui CLI.

Community
Advanced
Flux-FrontiersFlux-Frontiers
1

PyCodeKG Thorough Repository Analysis Skill

Analyze Python repository architecture and generate Markdown and JSON reports.

Official
Advanced
Arx-GameArx-Game
4

verify-against-code

Inspect codebases to classify surfaces as BUILT, NOT WIRED, or ABSENT.

Official
Intermediate
taylrfnttaylrfnt

code-archaeology

Index and query codebase knowledge graphs using Python and JSONL.

Community
Advanced