Skip to main content

Accord Engine

Accord Engine is a powerful Multi-Agent Decision Intelligence framework designed by Arclane Systems. It provides a robust, state-of-the-art infrastructure for executing and orchestrating complex negotiations between autonomous AI agents.

Whether you're building automated B2B procurement systems, autonomous trip planners, or dynamic resource allocators, Accord handles the hard parts of agent interaction: policy enforcement, cryptographically verifiable trust trails, long-term memory, and observability.

Features

  • Multi-Agent Orchestration: Seamlessly run complex, multi-round negotiations between customized agents (powered by LangGraph).
  • LLM-Driven Policies: Agents negotiate intelligently using Groq-powered LLMs, balancing strict priorities and flexibilities.
  • Persistent Agent Memory: Deep memory integration via Letta ensures agents remember past outcomes and opponents across sessions.
  • Cryptographic Trust: Every offer and counter-offer is signed and verified using asymmetric cryptography for complete auditability.
  • Deep Observability: Tracing via Langfuse provides granular visibility into every LLM call and agent reasoning step.
  • Rich Reporting: Automatically generate beautiful HTML negotiation transcripts and visual trace reports.

Quick Start

pip install accord-engine
from accord import NegotiationSession, Agent, Preference

# 1. Define your agents and their hidden preferences
buyer = Agent(
    name="ProcurementBot",
    preferences=[
        Preference(name="budget", value=5000, priority=0.9, flexibility=0.2),
        Preference(name="timeline", value=30, priority=0.6, flexibility=0.5),
    ]
)

seller = Agent(
    name="VendorBot",
    preferences=[
        Preference(name="budget", value=7500, priority=0.9, flexibility=0.3),
        Preference(name="timeline", value=45, priority=0.4, flexibility=0.8),
    ]
)

# 2. Spin up a secure negotiation session
session = NegotiationSession(
    agents=[buyer, seller], 
    max_rounds=15
)

# 3. Let them negotiate
result = session.run()

print(f"Outcome: {result.outcome}")
if result.success:
    print(f"Final Agreed Terms: {result.final_terms}")

print(f"Reasoning: {result.reasoning}")

Documentation

Full documentation, architecture references, and advanced usage guides are available at Arclane Systems.


Built with precision for the next generation of autonomous infrastructure.

Metadata

Release files for accord-engine 1.0.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for accord-engine 1.0.1
File Size Uploaded Provenance Actions
accord_engine-1.0.1.tar.gz 32.0 kB Without provenance Details for accord_engine-1.0.1.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for accord-engine 1.0.1
File Interpreter ABI Platform Actions
accord_engine-1.0.1-py3-none-any.whl Python 3 none any Details for accord_engine-1.0.1-py3-none-any.whl

Total release size: 73.7 kB

Release files / accord_engine-1.0.1.tar.gz

Download URL accord_engine-1.0.1.tar.gz
Size 32.0 kB
Tags Source
SHA-256 checksum
How to use checksums
0b28e40ac6ac4661b9c1d951fdd7db3d1a2dedd8066cb6dc12e53406ab30adac
BLAKE2b-256 checksum
How to use checksums
f8e952d9ebe9aea84bd776979ddf67e096d63b0795391a6e7a5e98e26abc4471
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.4

Release files / accord_engine-1.0.1-py3-none-any.whl

Download URL accord_engine-1.0.1-py3-none-any.whl
Size 41.7 kB
Tags Python 3
SHA-256 checksum
How to use checksums
dcd0056b92deb2e7fdf25b50ff58e46e3731b79cd01a48f80abbcc091e6d90d8
BLAKE2b-256 checksum
How to use checksums
603cf57ae078da60832c0079da32fc834ca35c62773f2330465dc756fe3ff3de
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.4

Security for accord-engine 1.0.1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

1.0.1 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page