maigp-crewai

AIGP governance for CrewAI multi-agent orchestration.

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 CrewAI crews, tasks, and agents.

Governance model

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

Install

pip install maigp-crewai

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

Quick start

from crewai import Crew
from aigp_crewai import AigpCrewCallbacks

callbacks = AigpCrewCallbacks(gov_url="https://gov.example.com",
                              app_id="my-crew", hmac_secret="secret")

# Attach to your Crew so each task runs as a governed child session
crew = Crew(agents=[...], tasks=[...], step_callback=callbacks.on_step)

Multi-agent model

Crew start opens a parent session (CHECK); each task runs as a child session with a delegation token; each agent model call accumulates tokens; crew end emits an aggregate RECORD with linked TRACEs.

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.