llamaindex_agentmesh 3.5.0

AgentMesh trust layer integration for LlamaIndex agents

LlamaIndex AgentMesh IntegrationAgentMesh trust layer integration for LlamaIndex - enabling cryptographic identity verification and trust-gated agent workflows.## OverviewThis integration provides:- TrustedAgentWorker: Agent worker with cryptographic identity and trust verification- TrustGatedQueryEngine: Query engines with access control based on trust- Secure Data Access: Governance layer for RAG pipelines with identity-based policies## Installationbashpip install llama-index-agent-agentmesh## Quick Start### Creating a Trusted Agentpythonfrom llama_index.agent.agentmesh import TrustedAgentWorker, VerificationIdentity# Generate cryptographic identityidentity = VerificationIdentity.generate( agent_name="research-agent", capabilities=["document_search", "summarization"],)# Create trusted agent workerworker = TrustedAgentWorker.from_tools( tools=[search_tool, summarize_tool], identity=identity, llm=llm,)# Create agent with trust verificationagent = worker.as_agent()### Trust-Gated Query Enginepythonfrom llama_index.agent.agentmesh import TrustGatedQueryEngine, TrustPolicy# Wrap query engine with trust policytrusted_engine = TrustGatedQueryEngine( query_engine=base_engine, policy=TrustPolicy( min_trust_score=0.8, required_capabilities=["document_access"], audit_queries=True, ),)# Query requires verified identityresponse = trusted_engine.query( "What are the quarterly results?", invoker_card=requester_card,)### Multi-Agent Trust Handoffspythonfrom llama_index.agent.agentmesh import TrustHandshake, TrustedAgentCard# Create agent card for discoverycard = TrustedAgentCard( name="research-agent", description="Performs document research", capabilities=["search", "summarize"], identity=identity,)card.sign(identity)# Verify peer before task handoffhandshake = TrustHandshake(my_identity=identity)result = handshake.verify_peer(peer_card)if result.trusted: # Safe to delegate task pass## Features### TrustedAgentWorkerAn agent worker that:- Has cryptographic identity for authentication- Verifies peer agents before accepting tasks- Signs outputs for verification by recipients- Supports capability-based access control### TrustGatedQueryEngineA query engine wrapper that:- Requires identity verification for queries- Enforces trust score thresholds- Restricts access based on capabilities- Provides audit logging of all queries### Data Access GovernanceControl access to your RAG pipeline:pythonfrom llama_index.agent.agentmesh import DataAccessPolicypolicy = DataAccessPolicy( allowed_collections=["public", "internal"], denied_collections=["confidential"], require_audit=True, max_results_per_query=100,)# Apply policy to indextrusted_index = TrustedVectorStoreIndex( index=base_index, policy=policy,)## Security ModelAgentMesh uses Ed25519 cryptography for:- Identity Generation: Unique DID per agent- Request Signing: All queries are signed- Response Verification: Outputs can be verified## API Reference| Class | Description || ----------------------- | ------------------------------------ || VerificationIdentity | Cryptographic agent identity || TrustedAgentWorker | Agent worker with trust verification || TrustGatedQueryEngine | Query engine with access control || TrustHandshake | Peer verification protocol || TrustedAgentCard | Agent discovery card || DataAccessPolicy | RAG access governance |## LicenseMIT License