MELRA
Local runtime that checks agent tool calls against policy, verifies evidence, and records receipts.
Tested on Linux
Use it
Linux x86_64
You need first
- Runtimes:
Node.js 22.13 or newer, or 23.4 or newer,Python 3.11 to 3.14
Install MELRA CLI globally via npm
npm install -g @melra/cli@alpha
Add it to your agent
claude mcp add melra -- melra serveJSON for Claude Desktop, Cursor and other clients
{
"mcpServers": {
"melra": {
"command": "melra",
"args": [
"serve"
]
}
}
}Add MELRA as an MCP server in your MCP client, then call melra_capabilities to see what it offers.
What it does
- Check AI agent tool calls against policyNot shown working
- Require approval before risky actions runNot shown working
- Verify the results of tool callsNot shown working
- Avoid blind retries after timeoutsNot tested
How far pickfit got
Linux x86_64
- Installs
- Connects
- 0 of 4 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 31 Oct 2026
- Version 0.4.0-alpha.2
An MCP client handshake over stdio listed 12 tools, and a call to melra_capabilities returned a result.