AI agent passports define who can act and under what limits.
An AI agent passport records an agent's owner, capabilities, and limits. APort uses that identity record with policy checks to authorize each connected tool call before it runs. The passport declares authority; the installed guardrail enforces it.
Buyer
Who this is for
Platform and security teams that need to know which agent is acting, who owns it, and what it is allowed to do.
Outcomes
- Replace anonymous automation with named, owned, scoped agent identities.
- Keep your identity provider for login and delegated access; add policy checks at connected tool boundaries.
- Use the same identity primitive across GitHub, Claude Code, Cursor, MCP, and custom agents.
Controls
- Owner and organization binding
- Capability and limit declarations
- Assurance-level requirements
- Region and residency metadata
- Status and suspension controls
Proof
- Passport JSON
- Hosted passport record
- Capability-to-policy matching
- Signed decisions referencing the passport
Setup
Set up AI Agent Passports the way your team works.
Start agent-first when the agent has repo context, switch to human review when a security owner wants the path, or use the CLI when the integration is already clear.
Recommended
Let my agent prepare AI Agent Passports.
Use this when an agent needs an identity, owner, capabilities, and audit scope.
Create an APort AI agent passport for this agent. Read https://aport.io/skillCopy this into the agent that already has access to your repo or runtime environment. Require a diff before enforcement or secrets are enabled.
Implementation
From passport to an enforced decision
Define the owner and allowed capabilities in the builder, then install a guardrail in your runtime. A connected hook checks the passport and action against policy. Hosted verification can return a signed decision for audit. A passport alone does not intercept tools or replace your identity provider.
Install my passport guardrailRelated products
Build the control path in layers.
Runtime Enforcement
APort guardrails evaluate shell, file, web, MCP, repository, and framework tool calls before the agent can act.
AI Control Packs
AI Control Packs are APort's enterprise-facing policy bundles for finance, data, repository, messaging, web, system, MCP, and deliverable workflows.
Decision Audit Trails
APort records policy decisions with context, reasons, timestamps, and signatures so teams can reconstruct what happened and why.