Graph Build
- 73 installs
- 325 repo stars
- Updated August 2, 2026
- athola/claude-night-market
Graph-build is an agent skill that builds or updates a tree-sitter SQLite code knowledge graph at `.gauntlet/graph.db` for search and blast-radius analysis.
About
Graph-build is an agent skill for solo and indie builders who want structural awareness of a codebase without hand-exploring every file. It detects the working directory (or a path you specify), chooses a full build when `.gauntlet/graph.db` is missing or an incremental update when the database already exists, and runs the bundled Python script with `--incremental` when appropriate. The output is a SQLite knowledge graph derived from tree-sitter ASTs, capturing nodes such as files, classes, functions, types, and tests plus relationship edges across calls, imports, inheritance, containment, implementation, and test links. Use it at the start of a coding session, after large refactors, or whenever you ask about codebase layout before blast-radius or flow-tracing skills. It pairs naturally with search and impact-analysis workflows in the same plugin and keeps the agent grounded in multi-language repos spanning Python, TypeScript, Go, Rust, Java, and many others.
- Full or incremental builds of `.gauntlet/graph.db` via `graph_build.py`
- 20+ languages parsed into File, Class, Function, Type, and Test nodes
- Edges for CALLS, IMPORTS_FROM, INHERITS, CONTAINS, IMPLEMENTS, and TESTED_BY
- JSON report with files parsed, nodes, edges, and duration
- Suggests graph search and blast-radius analysis as follow-ups
Graph Build by the numbers
- 73 all-time installs (skills.sh)
- Ranked #5,587 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Security screen: LOW risk (skills.sh audit)
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/athola/claude-night-market --skill graph-buildAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 73 |
|---|---|
| repo stars | ★ 325 |
| Security audit | 3 / 3 scanners passed |
| Last updated | August 2, 2026 |
| Repository | athola/claude-night-market ↗ |
What it does
Build or refresh a local SQLite code knowledge graph so your agent can search structure and trace blast radius before risky edits.
Who is it for?
Multi-language codebases where you run agent sessions that depend on call graphs, imports, and test linkage before editing.
Skip if: Repos where you only need a one-off file read, have no Python 3 available to run the script, or do not use the Gauntlet graph workflow.
When should I use this skill?
Setting up the graph before search or blast-radius analysis; at session start; after significant code changes; when the user asks about codebase structure.
What you get
You get an up-to-date `.gauntlet/graph.db` with parsed node and edge counts so you can run graph search or blast-radius analysis on confident structural context.
- .gauntlet/graph.db knowledge graph
- JSON build summary (files, nodes, edges, duration)
By the numbers
- 20+ languages supported
- 5 node types (File, Class, Function, Type, Test)
- 6 edge relationship kinds
Files
Build Code Knowledge Graph
Build or update the .gauntlet/graph.db knowledge graph for the current codebase.
Steps
1. Detect target: Use the current working directory or a user-specified path.
2. Check for existing graph: If .gauntlet/graph.db exists, run an incremental update. Otherwise, run a full build.
3. Run the build script:
For full build:
python3 ${CLAUDE_PLUGIN_ROOT}/scripts/graph_build.py <dir>For incremental update:
python3 ${CLAUDE_PLUGIN_ROOT}/scripts/graph_build.py <dir> --incremental4. Report results: Show the JSON output including files parsed, nodes created, edges created, and duration.
5. Suggest next steps: Recommend searching the graph or running blast radius analysis.
When To Use
- At the start of a session to build structural awareness
- After significant code changes to update the graph
- Before running blast radius analysis or flow tracing
- When the user asks about codebase structure
What Gets Parsed
The graph extracts nodes (File, Class, Function, Type, Test) and edges (CALLS, IMPORTS_FROM, INHERITS, CONTAINS, IMPLEMENTS, TESTED_BY) from 20+ languages including Python, JavaScript, TypeScript, Go, Rust, Java, C/C++, Ruby, and PHP.
Storage
- Database:
.gauntlet/graph.db(SQLite with WAL mode) - Auto-creates
.gauntlet/.gitignoreto prevent commits - Incremental updates use SHA-256 hashing to skip
unchanged files
Exit Criteria
- [ ]
.gauntlet/graph.dbexists and is a valid SQLite file after
the skill completes; .gauntlet/.gitignore exists to prevent the database from being committed
- [ ] Build report JSON includes all four keys:
files_parsed,
nodes_created, edges_created, duration; values are non-zero for non-empty codebases
- [ ] Incremental update path used when
.gauntlet/graph.dbalready
exists (script invoked with --incremental); full build only on first run
- [ ] Next-step suggestions presented to the user: search the graph
or run blast-radius analysis
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10How it compares
Use this structured graph build instead of asking the agent to grep and guess relationships file by file.
FAQ
Who is graph-build for?
Developers using Claude Code or similar agents on real codebases who want AST-backed structural maps before search or impact analysis.
When should I use graph-build?
At session start for awareness, after significant code changes, before blast-radius or flow tracing, or when you ask how the codebase is organized—in Build agent-tooling and again in Ship review or Operate iteration when structure drifted.
Is graph-build safe to install?
It runs a local Python build script against your project tree; review the Security Audits panel on this Prism page and inspect `graph_build.py` before running on sensitive repos.