Show HN: Axtary – Content Authorization for AI Agents
6 hours ago
- Axtary authorizes AI agents by verifying exact action payloads (diff, message, query, tool payload) before execution.
- Routine actions follow deterministic policy; higher-risk actions require human approval of the exact payload, bound to its hash.
- The system records every attempt in a ledger; approval is tied to the payload, not a summary, preventing changes after review.
- Enforcement runs locally alongside the agent, keeping provider credentials local; the hosted control plane coordinates policies, approvals, and audits.
- Governed actions include code changes, infrastructure paths, data access, messages, tickets, documents, and MCP tools.
- ActionPass is a security artifact containing normalized action, payload hash, constraints, and policy, used across SDKs, proxies, and MCP wrappers.
- Policy can be expressed via Cedar/Rego languages or local configuration files; installation and demo are available via npm.