Strata
Shared memory for fleets of AI coding agents with LLM-judged contributions.
Tested on Linux
Use it
Linux x86_64
You need first
- Runtimes:
python>=3.11 - Keys or accounts: JUDGE_API_KEY
pipx install strata-mem
pipx install strata-mem
Check that it works
strata --version 2>&1The output should contain “strata 1.19.0”.
Add it to your agent
claude mcp add strata -- strata-mcpJSON for Claude Desktop, Cursor and other clients
{
"mcpServers": {
"strata": {
"command": "strata-mcp",
"args": []
}
}
}Run strata register --description "<your description>" inside a git repository to set up a fleet, then point your MCP client at strata-mcp.
What it does
- Share memory across a fleet of AI coding agentsTested
How to use it
- In a git repository, run
strata register --description "<your description>"to create the fleet config in.strata/. - Add
strata-mcpas an MCP server in your coding agent's client. - Call the
strata_list_scopestool to see the shared memory scopes.
- In a git repository, run
- Write learnings to a scope and read others' learningsNot shown working
- Have an LLM judge check every writeNot tested
How far pickfit got
Linux x86_64
- Installs
- Connects
- 1 of 3 features work
- Works offline
How pickfit tested it
Each run starts from a clean machine and follows the steps above. How pickfit verifies
Linux x86_64
An MCP client called one of its tools
- Replayed from scratch
- Works offline
- Checked on 1 Nov 2026
- Version 1.19.0
An MCP client handshake over stdio listed 17 tools, and a call to strata_list_scopes returned a result.