This MCP server provides cross-session memory and recall for AI coding assistants, allowing them to retain knowledge across different sessions and tools. It features a git-synced knowledge base and a TF-IDF ranked session search.
From the registry: Cross-session memory for AI agents - knowledge graph, scoring, semantic search

Please install the `agent-knowledge` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Node.js >= 20** — Required runtime (https://nodejs.org/)
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["agent-knowledge"]
- optional environment variables:
- `KNOWLEDGE_MEMORY_DIR`: Git-synced knowledge base directory (example: `~/agent-knowledge`)
- `KNOWLEDGE_GIT_URL`: Git remote URL (auto-clones if missing) (example: `<your-git-url>`)
- `KNOWLEDGE_AUTO_DISTILL`: Auto-distill session insights (example: `true`)
- `KNOWLEDGE_EMBEDDING_PROVIDER`: local|openai|claude|gemini (example: `local`)
- `KNOWLEDGE_ANTHROPIC_API_KEY`: API key for Claude/Voyage embeddings (example: `<your-anthropic-api-key>`)
- `KNOWLEDGE_PORT`: Dashboard HTTP port (example: `3423`)
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.EXTRA_SESSION_ROOTSComma-separated paths for additional session roots.KNOWLEDGE_INDEX_VERBATIMToggle for verbatim session indexing.AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
Persistent codebase knowledge graph. Survives session restarts and context compaction.
A powerful toolkit for coding, providing semantic retrieval and editing capabilities.