scan_security
Security vulnerability scan: 40+ dangerous function patterns, source-to-sink taint tracing, auth coverage for HTTP endpoints (7 languages/frameworks), architectural layer violations, weak crypto, hardcoded secrets
How to use it
scan_security is exposed by the CodeGraph MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the scan_security tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.
Install CodeGraph
npx @vscode/vsce package # VSIXOther tools in CodeGraph (39)
Cyclomatic complexity with breakdown (branches, loops, nesting, exceptions, early returns)
Module coupling metrics and instability scores
Blast radius prediction — what breaks if you modify, delete, or rename
8 — search + symbol info + AI context
Find identifiers described in docs that don't exist in code yet — build TODO lists from specs
Find files importing a module
Search by param count, return type, modifiers
Detect circular import/dependency chains across files
Find unused imports — modules imported but never referenced
Main functions, HTTP handlers, CLI commands, event handlers
Most-called functions ranked by transitive caller count
Find all functions registered as ops struct callbacks
Tests that exercise a given function
Auto-generate a structured ARCHITECTURE.md from the live code graph (modules, hot paths, complexity, circular deps)
Primary context tool. Intent-aware (explain/modify/debug/test) with token budgeting. Returns source, related symbols, imports, siblings, debug hints.
Function call chains (callers and callees)
Cross-codebase context for a natural language query ("how does auth work?")
File/module import relationships with depth control
Full symbol info: source, callers, callees, complexity
Everything needed before editing: source + callers + tests + memories + git history
High-level summary of a directory: file count, functions, language breakdown, top complex functions
Quick metadata: signature, visibility, kind
17 — callers/callees/deps/impact/traverse/PR context
Add directory to graph alongside existing data
Add/update specific files without full reindex
Index a local .md file (ARCHITECTURE.md, APIDESIGN.md, etc.) into the persistent docs store. Heading-tree chunking with leaf-node embeddings.
List all indexed source files
14 — codegraphmemory plus the docs tools
Get memories relevant to a file/function
One-call PR review. Runs git diff against base branch, finds changed functions in the graph, reports: blast radius (callers), test coverage + gaps, affected modules, diff-aware change classification (signature vs body), stale-doc warnings, complexity, commit-message hint, suggested reviewers from gi
Full or incremental workspace reindex
Remove all indexed chunks from a source file
Find functions that throw, catch, or handle specific error types
Regex search across function bodies, signatures, names, and docstrings
Semantic search over indexed docs — returns matching sections with heading-path breadcrumbs
pro security tools only (empty on community)
Find symbols by name or natural language (hybrid BM25 + semantic search)
Custom graph traversal with edge/node type filters
Cross-reference doc claims vs code graph. direction=forward (doc→code), reverse (code→doc), or both