About APort
AI agent passports, pre-action authorization, and proof
Our Mission
APort gives AI agents passports and checks proposed actions before tools run. Teams use it to control shell commands, file reads, MCP tools, repository changes, and other agentic work while keeping signed decisions for audit.
Secure
Signed passports and decisions make agent actions reviewable
Universal
Works across hosted verification, local JSON passports, and framework hooks
Developer-First
One-command setup for Claude Code, Cursor, GitHub, and agent runtimes
Key Features
Agent Passports
OAP passports define agent identity, owner, assurance level, capabilities, limits, regions, and status.
- Unique agent identifiers
- Capability and limit declarations
- Hosted and local passport modes
- Status and suspension controls
Fast Verification
Low-latency hosted verification with edge routing and optimized APIs for high-performance applications.
- Performance-sensitive verifier path
- Global CDN distribution
- Rate limiting protection
- Enterprise SLA by agreement
Easy Integration
Install guardrails for common agent workflows or call the hosted verifier directly.
- Claude Code and Cursor hooks
- GitHub Repository Guard
- MCP tool policy checks
- API and middleware paths
Trust Network
Publish verified agent profiles, trust badges, and evidence links when public discovery matters.
- Public agent directory
- Trust badges and widgets
- Signed verification evidence
- Claim flow for issued passports
Share the Trust
Copy a short note that points developers to APort without overstating claims.
This agent has an APort passport.
APort adds pre-action authorization, guardrails, and audit for AI agent tools.
Learn more: https://aport.io
By the Numbers
Ready to Get Started?
Start with one repository or one coding agent, then expand to hosted enforcement when the team needs shared policy and audit.