Agent Identity Management

Where the specs become real

The specifications are paper. OpenA2A AIM (Agent Identity Management) is the working steel, the reference platform for AIP and ATX behind a developer experience that fits on one line: secure("my-agent"). It mints identities, obtains ATX credentials from the OpenA2A Registry and verifies them, enforces authorization, and keeps its own audit log, so a developer gets the stack without assembling it by hand. Four pieces live elsewhere: ATX issuance, did:opena2a resolution and the ATP trust authority with its transparency log run in the OpenA2A Registry, and the AAP credential broker is a library in Secretless, not yet reachable from its shipped daemon (agent-authorization-protocol#1).

What AIM does at each layer

The developer workflow

From a developer's seat, the whole stack collapses into install, one call, and one decorator. Everything else happens underneath.

from aim_sdk import secure

agent = secure("weather-agent")          # identity + DID, one line

@agent.perform_action(capability="weather:fetch", risk_level="low")
def get_weather(city: str):
    return requests.get(API, params={"q": city}).json()

get_weather("San Francisco")
# → request signed, 5-step authorization runs, action logged,
#   trust score updated, all automatically

5-step fine-grained authorization

This is AIM's implementation of AAP's enforcement, run on every action. The first four steps each carry a latency budget under 10 ms in the AIM README; only the riskiest calls pay for the deepest check.

What's in the box

• Backend, the registration, authorization, trust-scoring and audit services.
• SDKs, Python, Java and TypeScript, with auto-detection for popular agent frameworks.
• Dashboard, agents, trust-score breakdowns, alerts, audit log, policies and webhooks.
• Observability, OpenTelemetry traces, metrics and logs, exported by the backend over OTLP/gRPC to a collector the operator configures (default target localhost:4317).