// [Code Intelligence] Use when you need to build or update the code review knowledge graph.
[HINT] Download the complete skill directory including SKILL.md and all related files
| name | graph-build |
| description | [Code Intelligence] Use when you need to build or update the code review knowledge graph. |
| version | 1.0.0 |
Goal: [Code Intelligence] Build or update the code review knowledge graph. Parses codebase with Tree-sitter into a structural graph (functions, classes, imports, calls, tests) stored in SQLite. Enables blast-radius analysis and graph-powered code review.
Workflow:
Key Rules:
file:line) with confidence >80% to act.Requires Python 3.10+ with: pip install tree-sitter tree-sitter-language-pack networkx
Check availability — Run via Bash:
python .claude/scripts/code_graph status --json
last_updated is null: graph never built → proceed with full buildlast_updated exists: graph exists → proceed with incremental updateBuild or update — Run via Bash:
python .claude/scripts/code_graph build --jsonpython .claude/scripts/code_graph update --jsonReport results from JSON output:
.code-graph/graph.db (SQLite, auto-gitignored)After build/update, graph connectors run automatically if configured in project-config.json:
graphConnectors.apiEndpoints)graphConnectors.implicitConnections)See /graph-connect-api and .claude/docs/code-graph-mechanism.md for details.
The graph database includes optimized indexes created automatically on first build:
idx_nodes_name — fast node name lookups for searchidx_edges_kind_source — composite index for filtered edge queries (kind + source)idx_edges_kind_target — composite index for filtered edge queries (kind + target)These indexes are defined in the init schema and auto-create in any new project on first graph build.
After building, the CLI automatically runs:
project-config.json → graphConnectors.implicitConnections[]This creates edges for MESSAGE_BUS, TRIGGERS_EVENT, PRODUCES_EVENT, TRIGGERS_COMMAND_EVENT, and API_ENDPOINT — enabling full system flow tracing via the trace command.
Run python .claude/scripts/code_graph describe --json to get MCP-style structured descriptions of all available CLI commands, their parameters, and usage. Useful for AI agents to discover graph capabilities programmatically.
build, update, status, blast-radius, query, connections, trace, search, find-path, batch-query, sync, export, export-mermaid, connect-api, connect-implicit, review-context, describe
| Invalid Command | Correct Alternative |
|---|---|
incremental | update --json (incremental is the default behavior of update) |
update --files <list> | update --json (auto-detects changed files via git diff) |
build --files <list> | build --json (always does full rebuild) |
sync --files <list> | sync --json (auto-detects from git) |
file_summary | connections <file> --json |
Build or incrementally update the persistent code knowledge graph for this repository.
AI Mistake Prevention — Failure modes to avoid on every task:
Check downstream references before deleting. Deleting components causes documentation and code staleness cascades. Map all referencing files before removal. Verify AI-generated content against actual code. AI hallucinates APIs, class names, and method signatures. Always grep to confirm existence before documenting or referencing. Trace full dependency chain after edits. Changing a definition misses downstream variables and consumers derived from it. Always trace the full chain. Trace ALL code paths when verifying correctness. Confirming code exists is not confirming it executes. Always trace early exits, error branches, and conditional skips — not just happy path. When debugging, ask "whose responsibility?" before fixing. Trace whether bug is in caller (wrong data) or callee (wrong handling). Fix at responsible layer — never patch symptom site. Assume existing values are intentional — ask WHY before changing. Before changing any constant, limit, flag, or pattern: read comments, check git blame, examine surrounding code. Verify ALL affected outputs, not just the first. Changes touching multiple stacks require verifying EVERY output. One green check is not all green checks. Holistic-first debugging — resist nearest-attention trap. When investigating any failure, list EVERY precondition first (config, env vars, DB names, endpoints, DI registrations, data preconditions), then verify each against evidence before forming any code-layer hypothesis. Surgical changes — apply the diff test. Bug fix: every changed line must trace directly to the bug. Don't restyle or improve adjacent code. Enhancement task: implement improvements AND announce them explicitly. Surface ambiguity before coding — don't pick silently. If request has multiple interpretations, present each with effort estimate and ask. Never assume all-records, file-based, or more complex path.
Critical Thinking Mindset — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.
MUST ATTENTION apply critical thinking — every claim needs traced proof, confidence >80% to act. Anti-hallucination: never present guess as fact.
MUST ATTENTION apply AI mistake prevention — holistic-first debugging, fix at responsible layer, surface ambiguity before coding, re-read files after compaction.
TaskCreate BEFORE startingfile:line evidence for every claim (confidence >80% to act)[TASK-PLANNING] Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using TaskCreate.