Skip to main content

MAP (Model Action Protocol) — verifiable provenance, self-healing critique, and state rollback for autonomous AI agents

Project description

model-action-protocol (Python)

Python implementation of the Model Action Protocol. Cryptographic provenance, self-healing, and state rollback for autonomous AI agents.

This package conforms to MAP spec version 0.1.0. Both this implementation and the TypeScript reference implementation produce byte-identical hashes for the same inputs, verified by the shared conformance fixtures in ../spec/fixtures/v0.1/.

Install

pip install model-action-protocol

Optional extras:

pip install "model-action-protocol[anthropic]"   # LLM critic via Anthropic SDK
pip install "model-action-protocol[sqlite]"      # async SQLite store
pip install "model-action-protocol[postgres]"    # async Postgres store
pip install "model-action-protocol[fastapi]"     # FastAPI demo deps
pip install "model-action-protocol[all]"

Quickstart

See examples/quickstart.ipynb for the full Jupyter walkthrough and examples/fastapi_app/ for an HTTP service template.

Specification

The wire format, canonicalization rule (RFC 8785), hash inputs, and chain verification algorithm are defined in spec/SPEC.md. Implementation details (Pydantic models, async I/O patterns, anthropic SDK integration) are not part of the spec — they are this implementation's choices.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

model_action_protocol-0.1.1.tar.gz (63.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

model_action_protocol-0.1.1-py3-none-any.whl (35.2 kB view details)

Uploaded Python 3

File details

Details for the file model_action_protocol-0.1.1.tar.gz.

File metadata

  • Download URL: model_action_protocol-0.1.1.tar.gz
  • Upload date:
  • Size: 63.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.3

File hashes

Hashes for model_action_protocol-0.1.1.tar.gz
Algorithm Hash digest
SHA256 76fcdad399ed7593fe8103f96289d5be7588782a8895fe06fa15d29d500ff143
MD5 bcdc61338bb881be4aba0960eb4d2506
BLAKE2b-256 87a341707b11a21eab4c50d19498ff7e21a817cf7b3f6395f116f7b49d660a9c

See more details on using hashes here.

File details

Details for the file model_action_protocol-0.1.1-py3-none-any.whl.

File metadata

File hashes

Hashes for model_action_protocol-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 da321a5816f93d176a685532841a75ab45603c4eea79adc2a45a64e3e34354e4
MD5 97d21236b184dafea52a9b5b7c2b2eb8
BLAKE2b-256 b55c6a16fa6b108bc485b3ee44d525367b2639ba369eef96b65abed034450ab3

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page