What problem does it solve?
Deep exploration of Forgetful's knowledge graph helps you uncover hidden context, understand how memories, entities, and relationships interconnect, and avoid missed connections when planning across topics.
Core Features & Use Cases
- Forgetful stores knowledge as an interconnected graph: memories link to other memories, entities link to memories, and entities relate to each other. Deep exploration reveals context that simple queries miss.
- Exploration Phases: Phase 1 through Phase 5 outline sequential steps to discover memories, expand details, discover entities, map relationships, and link entity memories for comprehensive context.
- Presenting Results: Group findings by Memories, Entities, Artifacts; provide a Graph Summary and depth control options (Shallow, Medium, Deep) for task-specific context.
- Efficiency Tips: Use query_memory results and filters to limit scope and avoid diminishing returns.
Quick Start
Start by querying a topic semantically to gather primary and linked memories, then progressively expand with get_memory, list_entities, get_entity_relationships, and get_entity_memories as needed. Example: execute_forgetful_tool("query_memory", {"query": "<topic>", "query_context": "Exploring knowledge graph for comprehensive context", "k": 5, "include_links": true, "max_links_per_primary": 5}) Then fetch details for key memories with: execute_forgetful_tool("get_memory", {"memory_id": <id>}) Then discover entities with: execute_forgetful_tool("list_entities", {"project_ids": [<ids>]}) Then map relationships with: execute_forgetful_tool("get_entity_relationships", {"entity_id": <id>, "direction": "both"}) Then fetch entity memories with: execute_forgetful_tool("get_entity_memories", {"entity_id": <id>})
