# mpl — Meaning Protocol Layer

> Contracts, quality measurement, and audit trails for AI agent communication.
> A Skelf-Research project. MIT-licensed. Repo: https://github.com/Skelf-Research/mpl

MPL is a protocol layer that sits between AI agents and their tools (MCP, A2A,
HTTP). It validates payloads against versioned semantic-type contracts, scores
them against configurable quality profiles (QoM), and writes BLAKE3-hashed,
provenance-attributed audit records.

## Core pages
- [Home](https://mpl.skelfresearch.com/): overview, stack diagram, two-minute start.
- [About](https://mpl.skelfresearch.com/about/): premise, what's in the repo, status, audience.
- [Blog](https://mpl.skelfresearch.com/blog/): notes on contracts, quality, and audit forensics.
- [Compare](https://mpl.skelfresearch.com/compare/): MPL vs MCP, MPL vs A2A.

## Blog posts
- [Why agent-to-agent calls need contracts, not just JSON](https://mpl.skelfresearch.com/blog/contracts-not-just-json/)
- [Quality measurement vs guardrails: a different layer of the stack](https://mpl.skelfresearch.com/blog/quality-vs-guardrails/)
- [Reading mpl audit trails: forensic patterns](https://mpl.skelfresearch.com/blog/reading-audit-trails/)

## Comparisons
- [mpl vs MCP](https://mpl.skelfresearch.com/compare/mcp/)
- [mpl vs A2A](https://mpl.skelfresearch.com/compare/a2a/)

## External
- Documentation: https://docs.skelfresearch.com/mpl/
- Source: https://github.com/Skelf-Research/mpl
- License: MIT