Metadata-Version: 2.4
Name: agent-coherence
Version: 0.1.0
Summary: MESI-style artifact coherence for multi-agent AI systems
Author: Arbiter contributors
License: Apache-2.0
Project-URL: Homepage, https://github.com/hipvlady/agent-coherence
Project-URL: Repository, https://github.com/hipvlady/agent-coherence
Project-URL: Issues, https://github.com/hipvlady/agent-coherence/issues
Keywords: multi-agent,llm,cache-coherence,mesi,token-efficiency,agents
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Requires-Python: >=3.11
Description-Content-Type: text/markdown
Requires-Dist: pyyaml>=6.0
Provides-Extra: dev
Requires-Dist: pytest>=7.0; extra == "dev"

# agent-coherence — The Coherence Protocol for AI Agents

[![CI](https://github.com/hipvlady/agent-coherence/actions/workflows/ci.yml/badge.svg)](https://github.com/hipvlady/agent-coherence/actions/workflows/ci.yml)

`agent-coherence` implements MESI-style cache coherence for shared artifacts in multi-agent LLM systems, reducing synchronization token overhead and preventing stale-context coordination failures.

## Install

```bash
pip install agent-coherence
```

## Quick start

```python
from ccs.simulation.engine import run_strategy_comparison
from ccs.simulation.scenarios import load_scenario

scenario = load_scenario("benchmarks/scenarios/planning_canonical.yaml")
report = run_strategy_comparison(scenario, strategies=["eager", "lazy"], runs=5, seed_start=20260305)
print(report.to_dict()["aggregated"])
```

## Reproduce benchmark artifacts

```bash
bash reproduce.sh
```

See [REPRODUCE.md](REPRODUCE.md) for full output mapping and baseline verification details.

## Paper

Token Coherence: MESI-Style Cache Coherence for Shared Artifacts in Multi-Agent LLM Systems

## License

Apache-2.0 (`LICENSE`)
