maigp-anthropic

AIGP governance for direct Anthropic Messages API calls.

Part of MAIGP — the Mediated AI Governance Protocol: a runtime consent / scope / evidence layer for AI systems. Every governed call is CHECKed before it runs (policy, scope, jurisdiction, circuit-breaker) and RECORDed + TRACEd as tamper-evident evidence after.

What it governs

Governs raw anthropic client calls (Messages API).

Governance model

This adapter maps the framework's lifecycle onto the AIGP agent-governance loop provided by maigp-agent-core:

Install

pip install maigp-anthropic

Pulls in maigp-agent-core (shared lifecycle) and, transitively, maigp-client.

Quick start

from anthropic import Anthropic
from aigp_anthropic import GovernedAnthropic

governed = GovernedAnthropic(client=Anthropic(),
                             gov_url="https://gov.example.com",
                             app_id="my-anthropic-app", hmac_secret="secret")
response = governed.message("claude-sonnet-4", "Hello", system="You are helpful")

Raw API vs agent loop

This wraps single Messages API calls. To govern the Claude Agent SDK agent loop (tool-use turns), use maigp-claude-agent-sdk.

Links

• Protocol, specs & full adapter catalog: https://github.com/owner-spec/aigp-protocol
• Issues: https://github.com/owner-spec/aigp-protocol/issues

License: Proprietary. © Kanjani AI Research & Causum.