Skip to main content

Verifiable task delegation between AI agents

Project description

TrustHandoff

Install:

pip install trusthandoff

TrustHandoff is a lightweight SDK for verifiable task delegation between AI agents.

It provides a canonical structure for transferring tasks between agents with:

  • identity
  • permissions
  • bounded execution
  • provenance
  • cryptographic signatures

TrustHandoff aims to become a secure delegation layer for multi-agent systems.


Why TrustHandoff exists

Current agent ecosystems solve only part of the problem:

Agent ↔ tools → MCP
Agent ↔ communication → A2A
Secure task delegation → missing

TrustHandoff fills this gap by defining a SignedTaskPacket that allows agents to safely hand off tasks.


Delegation flow

flowchart LR
    A[Planner Agent] -->|Create SignedTaskPacket| B[TrustHandoff Packet]
    B --> C[Research Agent]
    C --> D[Execute Task]

Installation

pip install trusthandoff


Example

from datetime import datetime, timedelta, timezone
from trusthandoff import SignedTaskPacket, Permissions

packet = SignedTaskPacket(
    packet_id="pk_001",
    task_id="task_001",
    from_agent="agent:planner",
    to_agent="agent:research",
    issued_at=datetime.now(timezone.utc),
    expires_at=datetime.now(timezone.utc) + timedelta(minutes=10),
    nonce="123",
    intent="Research company background",
    context={"company": "Example Corp"},
    permissions=Permissions(
        allowed_actions=["read", "search"],
        max_tool_calls=5
    ),
    signature_algo="Ed25519",
    signature="signature",
    public_key="public_key"
)

Example script

examples/example_agents.py

Run with:

python examples/example_agents.py


Core primitive

TrustHandoff defines a canonical task transfer structure:

SignedTaskPacket

This packet includes:

  • task identity
  • agent identity
  • permissions
  • constraints
  • provenance
  • cryptographic signature

Vision

TrustHandoff aims to become the trust layer for agent delegation in multi-agent systems.

Possible integrations:

  • LangGraph
  • AutoGen
  • CrewAI
  • OpenAI Agents
  • LlamaIndex
  • custom agent runtimes

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

trusthandoff-0.2.0.tar.gz (7.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

trusthandoff-0.2.0-py3-none-any.whl (6.9 kB view details)

Uploaded Python 3

File details

Details for the file trusthandoff-0.2.0.tar.gz.

File metadata

  • Download URL: trusthandoff-0.2.0.tar.gz
  • Upload date:
  • Size: 7.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.3

File hashes

Hashes for trusthandoff-0.2.0.tar.gz
Algorithm Hash digest
SHA256 839450749bbd2629ef00a1d8931799545a970fd7e523d6ffd5be4d9d1a41d80f
MD5 a549e8ca528724c9506487f74db8fbc0
BLAKE2b-256 743ed0deec07a8aeb585861fc935be13ecbb70a2a03e06086ca30b67ae40b2e9

See more details on using hashes here.

File details

Details for the file trusthandoff-0.2.0-py3-none-any.whl.

File metadata

  • Download URL: trusthandoff-0.2.0-py3-none-any.whl
  • Upload date:
  • Size: 6.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.3

File hashes

Hashes for trusthandoff-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 31b5df914776b6891ce1b169aea79aa4cfe3a3899ea1f0bbc039bbc01b067161
MD5 b5be274944832e52b5d197452260b444
BLAKE2b-256 ec988ec13a591feb3a2699964decb79a0aea9f41e5ca046612929aa6f0b6454d

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page