AIM provides cryptographic identity, capability-based access control, and continuous trust evaluation for autonomous AI agents. Secure your AI infrastructure with one line of code.
7-minute walkthrough
from aim_sdk import secure
# One line to secure any AI agent
agent = secure("my-agent")
# AIM automatically:
# ✓ Generates Ed25519 cryptographic identity
# ✓ Discovers and attests MCP servers
# ✓ Enforces capability-based access
# ✓ Evaluates trust score on each verification
# ✓ Logs all actions for auditAgent Identity Management (AIM) is the practice of creating, authenticating, authorizing, and governing digital identities for autonomous AI agents. Unlike traditional Identity and Access Management (IAM) designed for human users, agent identity management addresses the unique challenges of non-human, autonomous systems that make decisions and take actions without human intervention.
As AI agents become critical infrastructure—connecting to databases, calling APIs, accessing sensitive data, and making autonomous decisions—organizations need purpose-built security that goes beyond retrofitting legacy authentication systems.
| Aspect | Traditional IAM | Agent Identity Management |
|---|---|---|
| Identity Type | Human users | Autonomous AI agents |
| Decision Making | Human-in-the-loop | Autonomous, no human review |
| Behavior | Predictable workflows | Dynamic, emergent behavior |
| Permissions | Fixed roles (RBAC) | Capability-based, context-aware |
| Lifecycle | Long-lived accounts | Ephemeral, dynamic provisioning |
| Trust Model | Static trust after auth | Continuous trust evaluation |
AIM implements the five security principles that every autonomous AI agent needs
Every agent receives a unique Ed25519 keypair for unforgeable identity. No shared credentials, no API keys to rotate.
Agents are restricted to explicitly declared capabilities. Principle of least privilege enforced automatically.
8-factor trust scoring evaluates agent behavior and adjusts permissions based on evaluation results.
Every agent action is logged with cryptographic attribution for compliance and forensics.
Automatic discovery and verification of MCP servers your agents connect to.
Complete visibility into your AI agent infrastructure with dashboards and alerts.
AIM works as a local-only tool for individual developers or as a centralized server for teams managing fleets of agents. Start solo, upgrade to server mode when you need cross-machine governance.
Local, no server required
Identity stored in ~/.opena2a/aim-core/. Audit log as a local JSONL file, policies as YAML files. Zero infrastructure to manage. Ideal for individual developers securing their own agents.
npx opena2a-cli identity create --name my-agentCentralized fleet management
Deploy AIM Server, Dashboard, and PostgreSQL via Docker Compose. Centralized identity management with a REST API, OIDC token endpoint, and a fleet dashboard for managing agents across machines.
docker compose up -dSee the deployment documentation for detailed setup instructions.
ServiceNow disclosed what researchers called "the most severe AI-driven vulnerability to date"—exposing 85% of Fortune 500 companies through improperly secured AI agents with universal credentials, email-only authentication, and unrestricted capabilities.
Read the full analysisAIM secures AI agents across frameworks, platforms, and deployment models
Secure LangChain and LangGraph agent deployments
Identity management for multi-agent crews
Attest and monitor Claude Desktop MCP servers
Govern autonomous multi-agent workflows
Unlike proprietary solutions from Microsoft, SailPoint, or Okta, AIM is 100% open source. Self-host forever, audit every line of code, no vendor lock-in.
Apache-2.0 licensed. Inspect, modify, and contribute to the codebase.
Deploy on your infrastructure. Your data never leaves your environment.
No licensing fees, no per-agent pricing, no surprise bills.
AIM has three integration points. Most developers start with aim-core and add the server when they need centralized management.
opena2a identity
Create identities, calculate trust scores, manage governance, and view audit logs. The fastest way to get started with AIM. No server required.
npx opena2a identity create --name my-agentUse when:
@opena2a/aim-core
Embed identity directly into your application. Local-only library with Ed25519 keys, capability policies, audit logs, and trust scoring. No server required.
npm install @opena2a/aim-coreUse when:
opena2a/aim-server
Full server with REST API, dashboard, PostgreSQL-backed audit logs, OIDC-based identity, and centralized policy management across a fleet of agents.
docker pull opena2a/aim-serverUse when:
| Feature | aim-core (local) | AIM Server |
|---|---|---|
| Ed25519 Identity | Local keypair | Server-issued + OIDC |
| Audit Log | JSON-lines file | PostgreSQL + API |
| Capability Policy | YAML file | REST API + dashboard |
| Trust Scoring | 8-factor, local | Real-time + history |
| Multi-Agent | Per-machine only | Cross-machine fleet |
| Infrastructure | None | Docker / Kubernetes |
Deploy agent identity management in minutes. One line of code to secure any AI agent.
Open source • Apache-2.0 license • Self-hosted