Recall

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.

Install pip install ai-recallworks

How it works

Install

One pip install. Add 3 lines to your MCP config. No API key. No account. No Docker required.

pip install ai-recallworks

Remember

Your agent calls remember() as it works. Decisions, patterns, conventions, gotchas — all indexed locally with vector search.

Recall

Next session, the agent calls recall() and gets back semantically relevant context. No re-explaining. No scrollback. No cold start.

The market

40M+
developers using AI coding tools
GitHub, Google, JetBrains surveys 2025
100%
start every chat cold
No agent ships with persistent memory
$0
switching cost
MCP config change — keep your agent

vs. funded competitors

Mem0 (YC, cloud SaaS) Letta (Berkeley, agent platform) Recall
What it isCloud memory APIA whole agentMCP memory server
Where data livesTheir serversTheir appYour disk
Works withCustom apps via SDKLetta onlyAny MCP client
SetupAPI key + code changesSwitch agents entirely1 config line
CostPaid SaaSFree + paid tiersFree forever (MIT)
MCP-nativeNoNoYes
Multi-agent coordNoNo6 primitives
Offline / air-gapNoNoYes

Mem0 and Letta ask you to change how you work. Recall gives memory to the tools you already use. Full comparison →

Start here

What it solves

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.

Open source + business model

Free forever (MIT)

  • Full MCP memory server
  • All 22 tools including multi-agent coordination
  • Local vector search (ChromaDB)
  • Offline / air-gapped operation
  • BYO embedder + LLM
  • Append-only artifacts on disk
  • Single-user, single-machine

Recall Cloud (coming)

  • Hosted brain with cross-machine sync
  • Team namespaces — shared memory across developers
  • Backup + retention policies
  • Admin dashboard + audit logs
  • SSO / SAML for enterprise
  • Priority support + SLA
  • From $99/mo per team

The free tier is genuinely great, forever. The paid tier sells convenience and team features — not unlocked functionality.

By the numbers

22
MCP tools
1
pip install
0
API keys required
MIT
License (core)