# melra

> melra — xagi-lab-melra. Use this tool when you need to ensure reliable and governed autonomous execution in various applications. MELRA solves problems related to durability and verifiability of autonomous systems by providing a modular execution layer for agents. It takes in model outputs and provides a runtime environment for secure and trustworthy autonomous decision-making.

Canonical page: https://skillsregistry.net/skills/xagi-lab-melra  
JSON: https://api.skillsregistry.net/v1/skills/xagi-lab-melra

## Description

The open-source autonomy kernel MELRA — Modular Execution Layer for Reliable Autonomy. An agent-independent effect runtime for governed, durable, and verifiable autonomous execution. MELRA begins where the tool call leaves the model loop.

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** container
- **Runtime environment:** vm
- **License:** Apache-2.0
- **Updated:** 2026-09-23

## Source

- **Source listing:** [GitHub](https://github.com/XAGI-Lab/melra)

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "xagi-lab-melra"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/xagi-lab-melra` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/xagi-lab-melra/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt