Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
RBT_ROOT_DIRYesThe path to your RBT documents directory

Tools

Functions exposed to the LLM to take actions

NameDescription
search_memory_nodes
Search the graph memory for relevant node summaries. These contain a summary of all of a node's relationships with other nodes. Note: entity is a single entity type to filter results (permitted: "Preference", "Procedure"). Args: query: The search query group_ids: Optional list of group IDs to filter results max_nodes: Maximum number of nodes to return (default: 10) center_node_uuid: Optional UUID of a node to center the search around entity: Optional single entity type to filter results (permitted: "Preference", "Procedure") Returns: List of node dictionaries containing search results Example: search_memory_nodes( query="project architecture decisions", group_ids=["knowledge-smith"], max_nodes=5 ) @REQ: REQ-graphiti-chunk-mcp @BP: BP-graphiti-chunk-mcp @TASK: TASK-007-MCPTools
search_memory_facts
Search the graph memory for relevant facts. Args: query: The search query group_ids: Optional list of group IDs to filter results max_facts: Maximum number of facts to return (default: 10) center_node_uuid: Optional UUID of a node to center the search around Returns: List of fact dictionaries containing search results Example: search_memory_facts( query="implementation dependencies", group_ids=["knowledge-smith"], max_facts=10 ) @REQ: REQ-graphiti-chunk-mcp @BP: BP-graphiti-chunk-mcp @TASK: TASK-007-MCPTools
get_episodes
Get the most recent memory episodes for a specific group. Args: group_id: ID of the group to retrieve episodes from. If not provided, uses the default group_id. last_n: Number of most recent episodes to retrieve (default: 10) Returns: List of episode dictionaries Example: get_episodes(group_id="knowledge-smith", last_n=5) @REQ: REQ-graphiti-chunk-mcp @BP: BP-graphiti-chunk-mcp @TASK: TASK-007-MCPTools
delete_episode
Delete an episode from the graph memory. Args: uuid: UUID of the episode to delete Returns: Success message dictionary Example: delete_episode(uuid="episode-uuid-123") @REQ: REQ-graphiti-chunk-mcp @BP: BP-graphiti-chunk-mcp @TASK: TASK-007-MCPTools
get_entity_edge
Get an entity edge from the graph memory by its UUID. Args: uuid: UUID of the entity edge to retrieve Returns: Entity edge dictionary containing edge details: { "uuid": "edge-uuid", "source_node_uuid": "source-uuid", "target_node_uuid": "target-uuid", "fact": "relationship description", "episodes": ["episode-uuid-1", "episode-uuid-2"], "valid_at": "2025-01-01T00:00:00Z", "invalid_at": null } Example: get_entity_edge(uuid="edge-uuid-123") @REQ: REQ-graphiti-chunk-mcp @BP: BP-graphiti-chunk-mcp @TASK: TASK-007-MCPTools
delete_entity_edge
Delete an entity edge from the graph memory. Args: uuid: UUID of the entity edge to delete Returns: Success message dictionary Example: delete_entity_edge(uuid="edge-uuid-123") @REQ: REQ-graphiti-chunk-mcp @BP: BP-graphiti-chunk-mcp @TASK: TASK-007-MCPTools
clear_graph
Clear all data from the graph memory and rebuild indices. WARNING: This operation is irreversible and will delete all data from the graph! Use with extreme caution. Returns: Success message dictionary Example: clear_graph() @REQ: REQ-graphiti-chunk-mcp @BP: BP-graphiti-chunk-mcp @TASK: TASK-007-MCPTools
add_document
Compare new file with ROOT original file and sync differences to Graphiti. This tool compares a modified document with its original version in the ROOT directory, chunks both versions, identifies changes, and syncs the differences to the Graphiti knowledge graph. Args: new_file_path: Absolute path to the new/modified file project_id: Project identifier (e.g., "knowledge-smith") feature_id: Feature identifier (required for RBT documents) rbt_type: RBT document type ("REQ"/"BP"/"TASK"). Leave as None for general documents. file_path: - For RBT TASK: task identifier (e.g., "006") - For general files: relative path (e.g., "todos/xxx.md" or "docs/todos/xxx.md") Note: "docs/" prefix is optional and will be handled automatically. Returns: Sync statistics: { "status": "success", "added": 3, # Number of chunks added "updated": 2, # Number of chunks updated "deleted": 1, # Number of chunks deleted "unchanged": 5, # Number of chunks unchanged "total": 11 # Total chunks } Raises: FileNotFoundError: If new file not found ValueError: If invalid rbt_type or parameter combination Examples: # RBT TASK document add_document( new_file_path="/Users/me/workspace/TASK-006-AddDocument.md", project_id="knowledge-smith", feature_id="graphiti-chunk-mcp", rbt_type="TASK", file_path="006" ) # RBT BP document add_document( new_file_path="/Users/me/workspace/BP-graphiti-chunk-mcp.md", project_id="knowledge-smith", feature_id="graphiti-chunk-mcp", rbt_type="BP" ) # General document (both work) add_document( new_file_path="/Users/me/workspace/TODO-001.md", project_id="General", file_path="todos/TODO-001.md" # or "docs/todos/TODO-001.md" ) @REQ: REQ-graphiti-chunk-mcp @BP: BP-graphiti-chunk-mcp @TASK: TASK-006-AddDocument

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/leo7nel23/KnowkedgeSmith-MCP'

If you have feedback or need assistance with the MCP directory API, please join our Discord server