Metadata-Version: 2.4
Name: teamrunner
Version: 0.0.1
Summary: The experimentation platform for agent organizations — run multi-agent orchestrations and factorial experiments entirely from text files.
Project-URL: Homepage, https://teamrunner.dev
Project-URL: Repository, https://github.com/teamrunner/teamrunner
Author-email: Kumar Muthuraman <kumar.muthuraman@gmail.com>
License-Expression: Apache-2.0
License-File: LICENSE
Keywords: agents,experimentation,llm,multi-agent,orchestration,reproducibility
Classifier: Development Status :: 1 - Planning
Classifier: Intended Audience :: Science/Research
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10
Description-Content-Type: text/markdown

# TeamRunner

**The experimentation platform for agent organizations.**

TeamRunner runs multi-agent orchestrations entirely from independent text files and folders — no coding required:

- **Agents** are Markdown files (config in frontmatter, system prompt as prose)
- **Teams** are folders that nest into hierarchies
- **Experiments** (A/B and full factorial) are declarative files
- Built-in logging, reproducibility manifests, and analytics make it a simulation lab for studying agent organizations — not just a workflow tool

Designed for researchers studying agentic behavior and organizational structures, and for eval/prompt-optimization teams.

## Status

🚧 **Pre-alpha.** The specification is being drafted and no runnable release exists yet. This package reserves the name and will carry the real implementation. Watch this repo for progress.

## License

Apache-2.0
