Give your AI agent a memory. One command. Zero config.
Every AI coding conversation starts cold. You re-explain the codebase, the conventions, the decisions — every time. Recall fixes that. An MCP memory server that works with Claude, Copilot, Cursor, and any MCP client. Local-first, MIT licensed, free forever.
pip install ai-recallworks
One pip install. Add 3 lines to your MCP config. No API key. No account. No Docker required.
pip install ai-recallworks
Your agent calls remember() as it works. Decisions, patterns, conventions, gotchas — all indexed locally with vector search.
Next session, the agent calls recall() and gets back semantically relevant context. No re-explaining. No scrollback. No cold start.
| Mem0 (YC, cloud SaaS) | Letta (Berkeley, agent platform) | Recall | |
|---|---|---|---|
| What it is | Cloud memory API | A whole agent | MCP memory server |
| Where data lives | Their servers | Their app | Your disk |
| Works with | Custom apps via SDK | Letta only | Any MCP client |
| Setup | API key + code changes | Switch agents entirely | 1 config line |
| Cost | Paid SaaS | Free + paid tiers | Free forever (MIT) |
| MCP-native | No | No | Yes |
| Multi-agent coord | No | No | 6 primitives |
| Offline / air-gap | No | No | Yes |
Mem0 and Letta ask you to change how you work. Recall gives memory to the tools you already use. Full comparison →
The full one-page treatment. Problem, six pillars, architecture diagrams, hardware reality, comparison to existing tools, cold-start protocol, where this matters.
What Agent OS replaces in the default workflow, and how it sits next to Cursor, Cline, Aider, Continue.dev, LangChain, AutoGen, and the rest.
Memory · Local compute · Vault · Guardrails · Brain backup · Continuity. Each one composable. None of them load-bearing alone.
Hot / warm / cold context tiers, the networked-brain pattern, and how a real session runs from trigger to overnight backup.
A vector database ready to scale to tens of thousands of chunks of your own knowledge. What it is, how to fill it, how to talk to it, and how it's different from /memories/.
What you need at each profile — workstation, mid-range laptop, integrated graphics, phone. Honest about what the GPU buys you.
Three commands sketch the cold-start protocol. Reference implementation links + scheduled-task layout.
Modern AI coding agents are stateless. Every conversation starts cold. Memory is what the host platform decides to keep. Costs are unbounded. Audit is whatever the chat transcript happens to retain. Multi-day campaigns degrade into "we already discussed this — read the scrollback" loops, until eventually the scrollback gets compacted by an opaque summarizer and detail vanishes.
Recall treats the agent like a contractor who clocks in every day: they don't remember yesterday's work from their own head, they read the project log. The log lives on disk, in a vector brain, and in immutable cloud storage. The agent's job is to keep that log honest and to read it before acting.
The free tier is genuinely great, forever. The paid tier sells convenience and team features — not unlocked functionality.