Metadata-Version: 2.5
Name: agentplay
Version: 0.1.0
Summary: Replay your coding-agent sessions (Claude Code, Codex, Cursor, Copilot, OpenCode, Kiro) and find what's wasting your agent's time. Local only.
Project-URL: Homepage, https://github.com/Entelligence-AI/agentplay
Project-URL: Issues, https://github.com/Entelligence-AI/agentplay/issues
Author: Entelligence AI
License-Expression: Apache-2.0
License-File: LICENSE
License-File: NOTICE
Keywords: claude-code,codex,coding-agent,copilot,cursor,developer-tools,observability
Classifier: Development Status :: 3 - Alpha
Classifier: Environment :: Console
Classifier: Intended Audience :: Developers
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Software Development :: Quality Assurance
Requires-Python: >=3.11
Provides-Extra: dev
Requires-Dist: pytest>=8; extra == 'dev'
Requires-Dist: ruff>=0.6; extra == 'dev'
Description-Content-Type: text/markdown

# agentplay

See what your coding agent actually did, find what's wasting its time, and fix your harness. Local, open source, and it works across Claude Code, Codex, Cursor, Copilot, OpenCode and Kiro.

```bash
uvx agentplay                 # build redacted replays of your local sessions, open the viewer
agentplay doctor --since 30d  # rank what's costing you retries, context and time
agentplay fix --list          # harness fixes: CLAUDE.md/AGENTS.md rules, allowlists, hooks
agentplay compare <recipe>    # did the fix help? before vs after, same metric
```

There are no runtime dependencies and no network calls, and nothing leaves your machine.

Full documentation: https://github.com/Entelligence-AI/agentplay
