# NathanNorman/claude-memory

Source: https://glimind.com/server/glama%2Fnathannorman%2Fclaude-memory
Registry id: `glama/nathannorman/claude-memory`
Status: unknown

## Tools (13)

- `codebase_search`: Semantic search over indexed source code repositories
- `community_search`: Identify tightly-coupled file clusters via Louvain community detection
- `dependency_search`: Query cross-repo dependency edges (repo_depends_on / repo_depended_on_by)
- `entity_browse`: List extracted entities (tools, projects, people) with occurrence counts
- `entity_graph`: Explore entity co-occurrence neighborhoods at depth 1-2
- `get_status`: Health check for both search backends with chunk/vector counts and model info
- `graph_traverse`: Walk upstream (callers) or downstream (callees) through the call graph
- `index_session`: Index a conversation session JSONL file (called by SessionEnd hook)
- `memory_deep_search`: 2-pass multi-hop retrieval: standard search, then entity extraction from top results seeds an expanded search
- `memory_read`: Read specific memory files or retrieve full past conversations by session UUID
- `memory_search`: Hybrid FTS5 keyword + vector cosine similarity search, merged via Reciprocal Rank Fusion (k=60)
- `memory_write`: Append to daily logs or long-term memory files, with immediate FTS5 indexing and vector embedding
- `symbol_search`: Find class/function/method definitions across indexed codebases (SQL LIKE patterns)

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.