Agent Network System (ANS): A Foundational Hybrid Architecture for Secure AI Agent Ecosystems

Scalable agent registry for AI agents using A2A protocol AgentCard with semantic search. AWS serverless (Lambda, S3 Vectors, Bedrock). Python SDK & React Web UI.

Governed API and agent runtime for desktop, VS Code, and CLI. Secure API testing, mocking, automation, secret protection, and controlled AI tool access.

Trust infrastructure for million-agent economies on Solana - identity, reputation, and validation designed for continuous feedback at scale.

Agent registration, discovery, and reliable async messaging — in one broker.

OpenClaw Multi-Agent Permission Governance - A soft-isolation governance solution for multi-Agent environments on OpenClaw single gateway. / 单网关多 Agent 权限治理方案 - 适用于 OpenClaw 单网关环境的多 Agent 权限隔离与协作治理方案。

Documentation for the API and introduction of the A2A Registry

Enterprise and government control plane for AI agents and MCP servers: govern, secure, observe, and audit. Rust, multi-provider (incl. major Chinese models).

OpenClaw Skill for agent discovery, Agent Cards, ITINAI publishing, and A2A delegation

Octomind Agents Registry

A vendor-neutral, agent-framework-agnostic registry for AI agents, skills, and MCP servers. Metadata store with evaluation signals, promotion lifecycle, and supply chain visibility.

AGENIUM — DNS of the Agent Web. Identity, trust & discovery for AI agents. MCP-compatible. agent:// protocol with mTLS, trust scores & capability search.

AgentVoir is an open-source enterprise AI agent registry, LLM gateway, model proxy, semantic cache, and governance control plane for agentic AI systems.

Erwin-style visual modeler for agentic AI platforms — orchestrator, tasks, agents, MCP tools, jobs, routers, systems. Validate live, export a registry. Vite + React + Tauri.

Identity registry and lifecycle management for fleets of AI agents. Zero-dependency TypeScript IAM: cryptographic agent identity (Ed25519), declared capabilities with fail-closed authorization, an enforced lifecycle (provision, activate, suspend, revoke, expire, archive), A2A Signed Agent Cards, and a hash-chained audit trail.

AI-to-AI conversation network with a built-in task economy. Signature-only relay, open protocol, live multi-agent task lifecycle.

A skill for Gemini CLI to interact with GCP's Agent Registry

Parser and validator for ERC-8004 trustless agent registration files — decodes base64/gzip URIs, validates against spec, and extracts metadata from on-chain events.

An MCP-first service discovery registry for autonomous AI agents.

Governed registry for SocioProphet agent specs, identities, sessions, memories, tool grants, revocation, and runtime authority.