mcp-knowledge-graph

What problem does it solve? Re-reading the same files every session wastes context and time. This Skill persists architectural facts about a codebase—components, dependencies, and decisions—into a queryable knowledge graph so you can recall project knowledge across sessions instead of re-analyzing the repository. ## Core Features & Use Cases - Persistent Entity Storage: Create typed entities (services, repositories, modules, decisions) with aim_memory_store and append observations with aim_memory_add_facts. - Relationship Mapping: Link entities with semantic relations like depends_on, calls, and implements using aim_memory_link. - Flexible Storage Locations: Choose project-local .aim/memory.jsonl storage for team knowledge or global stores for personal notes. - Use Case: After running repomix on a new repository, extract the key services and their dependencies, store them in the project graph, then query them with aim_memory_search in future sessions instead of re-reading source files. ## Quick Start Analyze this repository and store its main components, dependencies, and architectural decisions in the project knowledge graph for future sessions.