[Contrib] Agent-Mesh Trust Layer: Inter-Agent Trust Verification

[imran-siddique]

Summary

Adds inter-agent trust verification for MetaGPT multi-agent teams using Agent-Mesh.

Problem

MetaGPT enables powerful multi-agent collaboration, but agents interact without verifying trust:

• ProductManager sends requirements → Is PM authorized?
• Architect designs system → Can we trust the design?
• Engineer writes code → Should we execute untrusted code?

Solution

Trust verification at every interaction:

• TrustedRole: Wraps roles with cryptographic identity
• TrustPolicy: Configurable trust requirements
• TrustVerifier: Verifies interactions between agents
• TrustedTeam: Team wrapper with enforcement

Example

\\python
from metagpt.ext.agentmesh import TrustedTeam, TrustPolicy, TrustLevel

policy = TrustPolicy(
min_trust_level=TrustLevel.MEDIUM,
sensitive_actions={"WriteCode", "ExecuteCode"},
sensitive_action_trust=TrustLevel.HIGH,
)

team = TrustedTeam(policy=policy)
team.add_role(ProductManager(), trust_level=TrustLevel.HIGH)
team.add_role(Engineer(), trust_level=TrustLevel.MEDIUM)

Verifies trust before interaction

team.verify_message("ProductManager", "Engineer", "AssignTask")
\\

Changes

• Added \metagpt/ext/agentmesh/\
  • \ rust_layer.py\ - Core trust primitives
  • _init_.py\ - Public exports
  • \README.md\ - Documentation

Value for MetaGPT Users

Feature	Without Trust	With Agent-Mesh
Agent Identity	None	Cryptographic DIDs
Interaction Control	None	Policy-based
Sensitive Actions	Unrestricted	Trust-gated
Audit Trail	None	Full logging

References

• Agent-Mesh: https://github.com/imran-siddique/agent-mesh

[imran-siddique]

Ready for Final Review 🙏

This PR has been open for a while. The AgentMesh trust layer integration is complete and tested.

Could a maintainer please provide a final review? Happy to address any remaining concerns.

Thank you!

[imran-siddique]

Friendly nudge -- AgentMesh was just merged into microsoft/agent-lightning (14k stars) as part of Agent OS: microsoft/agent-lightning#478 -- Happy to address any feedback on this trust layer integration!

[imran-siddique]

Update: Our AgentMesh trust layer was just merged into LlamaIndex (47k stars): run-llama/llama_index#20644. This is our second major integration merge this week after Microsoft's agent-lightning (14k stars). Would love to get this PR reviewed as well!

[imran-siddique]

Friendly follow-up! Since opening this PR, the Agent-Mesh trust layer has been merged into three major frameworks:

• Dify (65K stars) -- langgenius/dify-plugins#2060 merged today
• LlamaIndex (47K stars) -- run-llama/llama_index#20644 merged
• Microsoft Agent-Lightning (15K stars) -- microsoft/agent-lightning#478 merged

Trust verification for MetaGPT's role-based teams (ProductManager -> Architect -> Engineer chain) is a natural fit. Happy to address any feedback.