Knowledge Graph MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| reindex_codebaseA | Scans and indexes the specified directory to build the knowledge graph. This is a heavy operation and should be called when the codebase changes significantly. Database will be created in {root_path}/.kgraph/ |
| search_codeA | Performs a semantic search for code snippets relevant to the query. Note: Requires a project to be indexed first with reindex_codebase. Parameters:
|
| get_structureB | Returns the structure (classes, functions) of a specific file. Note: Requires the file's project to be indexed first. |
| find_definitionsA | Finds where a class or function is defined. Returns file path, line number, and docstring. |
| find_referencesB | Finds where a symbol is used in the codebase. Returns a list of files and lines where the symbol is called. |
| get_file_summaryA | Returns a token-efficient summary of a file: imports, classes, functions, and docstrings. Optimized for small context windows. |
| get_usage_contextA | Retrieves a comprehensive context for a symbol:
This is optimized for small LLMs to get full context in one shot. |
| validate_editA | Validates a proposed code edit BEFORE applying it. Checks for:
Use this tool to check your code before writing it to a file. |
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 8 tools
Tools are mostly distinct, but get_structure, find_definitions, get_file_summary, and get_usage_context have overlapping information (e.g., both report classes/functions). Descriptions help differentiate, but an agent could be confused about which to use for a specific need.
All tools follow a consistent verb_noun pattern in snake_case (e.g., reindex_codebase, search_code, find_definitions). No mixing of styles or confusing verbs.
8 tools cover indexing, search, structural exploration, context retrieval, and validation—well-scoped for a code knowledge graph server. Each tool serves a distinct purpose without being too few or too many.
The surface covers core operations: indexing, search, definitions, references, file summary, and edit validation. Minor gaps exist (e.g., no explicit graph query tool), but the set is largely complete for code understanding tasks.