agentmemory is a local, persistent memory server for AI coding agents that automatically captures session events, compresses them into durable facts, and provides intelligent context retrieval.
Follow agentmemory to get an email when its listing changes, such as new pricing or features. One digest a day at most, and you can unsubscribe anytime.
It eliminates the need for manual re-explanation by allowing agents to recall project preferences, bugs, and architectures across different sessions and tools.
- 01
Automatic, zero-effort capture of agent tool usage and prompts via lifecycle hooks
- 02
Hybrid semantic search combining BM25, local vector embeddings, and knowledge graph traversal
- 03
Multi-agent support via MCP and REST API, enabling shared project memory across different tools
- 04
Memory evolution with versioning, auto-forgetting of stale facts, and contradiction detection
- 05
Real-time observation viewer and graph visualization on port 3113
Open Source
Open-source and free to self-host with local, offline embeddings. Optional cloud-based LLM features may incur usage costs from third-party providers.
Best for
Developers who frequently use multiple AI coding agents (such as Claude Code, Cursor, or Cline) on the same project and want to maintain long-term context without manual documentation.
Not ideal for
Users requiring a fully managed, cloud-hosted SaaS memory solution, as this tool is specifically designed for local-first, self-hosted deployment.
An essential productivity tool that effectively transforms static, per-session agent interactions into a cohesive, searchable project history with significant token savings.
Recent releases added Redis backend support, REST API authentication, improved graph compaction, and first-class support for GitHub Copilot CLI, Warp, and Zed, along with architectural fixes for durable capture and vector index stability.
As of
Is agentmemory free?+
Yes - agentmemory is Open Source. Open-source and free to self-host with local, offline embeddings. Optional cloud-based LLM features may incur usage costs from third-party providers.
Is agentmemory open source?+
Yes - agentmemory is open source. Open-source and free to self-host with local, offline embeddings. Optional cloud-based LLM features may incur usage costs from third-party providers.
Who is agentmemory best for?+
Developers who frequently use multiple AI coding agents (such as Claude Code, Cursor, or Cline) on the same project and want to maintain long-term context without manual documentation.
Who is agentmemory not ideal for?+
Users requiring a fully managed, cloud-hosted SaaS memory solution, as this tool is specifically designed for local-first, self-hosted deployment.
What are the best agentmemory alternatives?+
The closest agentmemory alternatives on ai.dosa.dev are Pieces for Developers, agenttrace, Vibe Kanban - all listed under Developer Productivity & Workflow.
Who makes agentmemory?+
agentmemory is developed by Rohit Ghumare. It is listed in the Developer Productivity & Workflow category on ai.dosa.dev.
Favorite this tool to revisit it later, or Zap it to contribute to the public vote count.