capiscio/a2a-demos

View on GitHub ↗

Demo agents showcasing CapiscIO Agent Guard and MCP Guard — trust badges, identity verification, and tool-level authorization for A2A and MCP protocols

23 ★0 forksPythonUpdated 3mo ago

What you need to know

Working demos of CapiscIO's 'Let's Encrypt for AI' trust infrastructure: cryptographic identity (DID), trust badges and policy enforcement for MCP servers and A2A agents, with LangChain/CrewAI/LangGraph agent demos.

Install

cd enforcement-demo && ./setup.sh
cd multi-agent-demo && ./setup.sh

Usage

  • •Follow the environment variable setup (CAPISCIO_API_KEY, CAPISCIO_SERVER_ID) and run the demo scripts
  • •Run the LangChain/CrewAI/LangGraph agents with python main.py --serve

Key features

  • ✓Server identity verification and per-tool min_trust_level
  • ✓Policy-as-code runtime enforcement without redeploys
  • ✓Badge verification, revocation and rate limiting
  • ✓A2A multi-agent demos with LangChain, CrewAI and LangGraph
  • ✓Registry, events and dashboard integration

Best for

Evaluating trust badges and policy enforcement for MCP/A2A setups

Caveats

  • ⚠Needs a free CapiscIO account and API key
  • ⚠Integration demos require an OpenAI API key
Platforms: macOS · Linux

Reviewed 2026-08-11

Topics

a2aa2a-protocolagent-to-agentai-agentsai-governanceai-securitydemosmcpmodel-context-protocolsecuritytrust-badges
Stars
23★
Forks
0
Language
Python
License
MIT
Created
2026-02-27
Last push
2026-05-22