code-context-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@code-context-mcpsearch code for 'payment processing'"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
code-context-mcp
MCP server for semantic code search and dependency graph analysis. Indexes codebases into a knowledge graph (Apache AGE) with vector embeddings (pgvector + Ollama) for AI-powered code understanding.
Supported Languages
TypeScript (
.ts,.tsx) — functions, classes, types, interfaces, importsRust (
.rs) — functions, structs, enums, traits, impl blocks, modules, use declarationsSolidity (
.sol) — contracts, functions, events, modifiers, structs, enums, interfaces, importsC - functions, enums
Related MCP server: code-graph-mcp
Prerequisites
Docker and Docker Compose
Node.js >= 18
Quick Start
# Start PostgreSQL (with AGE + pgvector) and Ollama
docker-compose up -d
# Install dependencies and build
npm install
npm run buildIndexing a Codebase
# Index a directory (project name auto-derived from directory basename)
node dist/cli/index.js index ./src
# Override the project name
node dist/cli/index.js index ./src --project my-project
# Index multiple codebases
node dist/cli/index.js index /path/to/frontend --project frontend
node dist/cli/index.js index /path/to/backend --project backendThe indexer will:
Parse all supported source files (TypeScript, Rust, Solidity) using tree-sitter
Build a code graph with functions, classes, types, contracts, traits, and their relationships
Generate vector embeddings for semantic search via Ollama
Show a progress bar during indexing
Re-running the indexer on the same directory skips unchanged files (detected via content hash).
Claude Desktop Configuration
Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"code-context": {
"command": "node",
"args": ["dist/mcp/server.js"],
"cwd": "/path/to/code-context-mcp",
"env": {
"PGHOST": "localhost",
"PGPORT": "5433",
"PGDATABASE": "rag_db",
"PGUSER": "postgres",
"PGPASSWORD": "postgres",
"OLLAMA_URL": "http://localhost:11434"
}
}
}
}MCP Tools
list_projects
List all indexed projects and their node counts.
Parameters: none
search_code
Semantic search across the indexed codebase. Finds functions, classes, types, and imports by meaning.
Parameter | Type | Required | Description |
| string | yes | Natural language search query |
| number | no | Maximum results (default: 10) |
| string | no | Scope search to a project |
get_call_stack
Trace the call dependency tree from a function.
Parameter | Type | Required | Description |
| string | yes | Function to trace from |
| number | no | Max traversal depth, 1-10 (default: 3) |
| string | no | Scope to a project |
get_function_context
Gather full context for a function: callers, callees, types it uses.
Parameter | Type | Required | Description |
| string | yes | Function to get context for |
| string | no | Scope to a project |
get_impact_analysis
Analyze impact of changes to a file. Shows all external code that depends on definitions in the file.
Parameter | Type | Required | Description |
| string | yes | Relative path of the file (as indexed) |
| string | no | Scope to a project |
Environment Variables
Variable | Default | Description |
|
| PostgreSQL host |
|
| PostgreSQL port |
|
| Database name |
|
| Database user |
|
| Database password |
|
| Ollama API endpoint |
Testing
Unit Tests
Unit tests run without Docker and cover extractors, scanner, registry, progress bar, embeddings, and DB queries.
npm test # Run all unit tests
npm run test:watch # Watch modeIntegration Tests
Integration tests require separate Docker containers to avoid interfering with development data.
# Start test infrastructure (separate ports: PG=5434, Ollama=11435)
docker compose -f docker-compose.test.yml up -d
# Run integration tests
npm run test:integrationIntegration tests cover the full indexing pipeline and multi-language graph population.
This server cannot be deployed
Maintenance
Related MCP Connectors
Repository knowledge graph MCP server for codebase understanding and debugging.
Hosted code graph over MCP: exact callers, dependencies, and cross-repo blast radius for AI agents.
An MCP server that gives your AI access to the source code and docs of all public github repos
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceAn MCP server that transforms codebases into intelligent, queryable knowledge bases, enabling AI assistants to perform semantic search, explore architecture, and analyze code relationships.167-
- AlicenseNot gradedqualityAmaintenanceA high-performance code knowledge graph server implementing MCP, indexing codebases into a structured AST knowledge graph with semantic search, call graph traversal, and HTTP route tracing.3,783 npm75MIT
- AlicenseAqualityFmaintenanceMCP server for semantic code search that indexes your codebase and allows AI editors to search using natural language queries.912 npm53MIT
- FlicenseNot gradedqualityBmaintenanceMCP server for AI coding agents that builds a complete code structure graph and semantic vector index, enabling fast querying of code entities, relationships, and impact analysis.11 npm10-