LangChain AgentMesh Integration
Cryptographic identity verification and trust-gated tool execution for LangChain agents.
Installation
pip install langchain-agentmesh
Features
- CMVKIdentity: Ed25519-based cryptographic identity for agents
- TrustGatedTool: Wrap any tool with trust requirements
- TrustedToolExecutor: Execute tools with automatic verification
- TrustCallbackHandler: Monitor trust events during chain execution
- TrustHandshake: Verify peer agents before collaboration
- DelegationChain: Hierarchical capability delegation
Quick Start
from langchain_agentmesh import CMVKIdentity, TrustGatedTool, TrustedToolExecutor
# Generate agent identity
identity = CMVKIdentity.generate('research-agent', capabilities=['search', 'summarize'])
# Wrap a tool with trust requirements
gated_tool = TrustGatedTool(
tool=search_tool,
required_capabilities=['search'],
min_trust_score=0.8
)
# Execute with verification
executor = TrustedToolExecutor(identity=identity)
result = executor.invoke(gated_tool, 'query')
Use Cases
Multi-Agent Trust Verification
from langchain_agentmesh import TrustHandshake
# Create handshake for peer verification
handshake = TrustHandshake(my_identity)
# Verify peer before collaboration
result = await handshake.verify_peer(peer_identity)
if result.trusted:
# Safe to delegate task
response = await peer_agent.invoke(task)
Trust-Gated Tool Execution
from langchain_agentmesh import TrustGatedTool
# Sensitive tool requiring high trust
code_execution_tool = TrustGatedTool(
tool=python_repl,
required_capabilities=['code:execute'],
min_trust_score=0.9,
audit_logging=True
)
# Only trusted agents can use this tool
result = executor.invoke(code_execution_tool, code)
Callback Integration
from langchain_agentmesh import TrustCallbackHandler
# Monitor trust events
callback = TrustCallbackHandler(
on_verification=lambda r: print(f"Verified: {r.peer_did}"),
on_violation=lambda v: alert(f"Violation: {v}")
)
agent = create_agent(callbacks=[callback])
Configuration
| Parameter | Default | Description |
|---|---|---|
min_trust_score |
0.5 | Minimum trust score required |
required_capabilities |
[] | Required capability list |
audit_logging |
False | Enable audit trail |
cache_ttl |
900 | Verification cache TTL (seconds) |
Related
License
MIT
Metadata
Release files for langchain-agentmesh 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| langchain_agentmesh-0.1.0.tar.gz | 13.9 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| langchain_agentmesh-0.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 29.0 kB
Release files / langchain_agentmesh-0.1.0.tar.gz
| Download URL | langchain_agentmesh-0.1.0.tar.gz |
|---|---|
| Size | 13.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
5819f4fced0c6ec183fa4f7c4a99c7b493256c31167fb0ea9603c1606110dbc6
|
|
BLAKE2b-256 checksum How to use checksums |
c874d3871fec2e5435ad0531cf4aaf23196c9851c14e5ac0bec8914cdc29c46a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Release files / langchain_agentmesh-0.1.0-py3-none-any.whl
| Download URL | langchain_agentmesh-0.1.0-py3-none-any.whl |
|---|---|
| Size | 15.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
a157a1a04b1576b20f3c1d28ae8ace82fb3a41583d3909283cd794e061a8a511
|
|
BLAKE2b-256 checksum How to use checksums |
1b3eb3fdf2b9f5293841404a0e8502b45fda2dea833425a58d5c44f2d465f9bd
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|