codegraph-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CODEGRAPH_DB | No | Path to the graph database file | graph.db |
| CODEGRAPH_TOKEN | Yes | Bearer token for authentication (generated via `codegraph token issue`) |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_symbolsB | Search indexed symbols (functions, methods, classes, types) by name substring. |
| get_symbolB | Fetch one symbol's full record (signature, location, container) by id. |
| find_referencesC | List every call site / identifier use of a name across the repo. |
| find_callersB | List symbols that directly call the given symbol (includes cross-language). |
| find_calleesB | List symbols the given symbol directly calls. |
| impact_analysisC | Transitive callers ('blast radius') of a symbol — what may break if it changes. |
| cross_language_edgesB | List resolved cross-language HTTP edges (e.g. a TS fetch -> a Go/Python handler). |
| find_orphansB | Dead-code candidates: functions/methods with no callers and not HTTP entrypoints. |
| find_hotspotsB | Most depended-on symbols (highest caller count) — where changes ripple furthest. |
| project_graphB | Module/package-level dependency graph: modules and the (weighted) edges between them. |
| graph_statsA | Summary counts for the indexed graph (files, symbols, edges, languages). |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 11 tools
Most tools target clearly distinct graph queries: search vs id-fetch, callers vs callees, transitive impact, orphan detection, hotspot ranking, module graph, and stats. The only soft overlap is find_references (raw call sites/identifier uses by name) versus find_callers (caller symbols, cross-language), which an agent could plausibly mix up despite the descriptions.
Ten of eleven names follow a predictable snake_case verb_noun pattern (search_symbols, get_symbol, find_callers, find_orphans, impact_analysis). The noun-phrase outliers cross_language_edges, project_graph, and graph_stats break the verb-first convention slightly, though they remain readable and self-describing.
Eleven tools is a well-scoped set for a code-graph analysis server; each tool covers a distinct analysis primitive (search, traversal, ranking, aggregation) with no filler. Nothing feels redundant or missing at the count level.
Coverage of the graph-analysis lifecycle is strong: discovery, symbol detail, reference/call traversal, transitive impact, cross-language edges, dead code, hotspots, module graph, and stats. Minor gaps remain — there is no tool to trigger/refresh indexing or to query a path between two symbols (shortest call path), which agents may expect.