agentplay is a local, open-source replay viewer for coding-agent sessions.
Follow agentplay to get an email when its listing changes, such as new pricing or features. One digest a day at most, and you can unsubscribe anytime.
It finds the Claude Code, Codex, Cursor, GitHub Copilot, OpenCode and Kiro transcripts already on your machine and turns each into a redacted map and playable timeline of what the agent read, edited, ran, retried and shipped.
- 01
One-command start with `uvx agentplay`, which discovers local agent sessions and opens a viewer on 127.0.0.1
- 02
Radial session map of files, tools, MCP servers, subagents and models touched
- 03
Replay timeline showing failed commands, retries, context-window fill and produced PRs or commits
- 04
Supports Claude Code (with subagent lanes), Codex, Cursor, Copilot CLI and VS Code, OpenCode and Kiro
- 05
Redacted map documents that omit prompt text, tool outputs, file contents and command arguments
- 06
No telemetry or uploads; the only network call is an optional email signup that can be disabled
Open Source
Free and open source under the Apache-2.0 license; run it with `uvx agentplay`, `pipx run agentplay` or `pip install agentplay`. No account or API key is required.
Best for
Developers and team leads who want to audit or debug what a coding agent actually did in a session, or share a sanitized session map with others.
Not ideal for
Teams wanting centralized, real-time observability across many machines; agentplay reads local transcripts on one machine, and fidelity for non-Claude agents is limited to the main lane.
A privacy-conscious, zero-setup way to see inside coding-agent sessions across the major harnesses; useful today, though still early.
Open-sourced by Entelligence AI on October 14, 2026 and published to PyPI (version 0.2.3 in the repository); it reached about 300 GitHub stars within ten days, with commits continuing on October 23, 2026.
As of
Is agentplay free?+
Yes - agentplay is Open Source. Free and open source under the Apache-2.0 license; run it with `uvx agentplay`, `pipx run agentplay` or `pip install agentplay`. No account or API key is required.
Is agentplay open source?+
Yes - agentplay is open source. Free and open source under the Apache-2.0 license; run it with `uvx agentplay`, `pipx run agentplay` or `pip install agentplay`. No account or API key is required.
Who is agentplay best for?+
Developers and team leads who want to audit or debug what a coding agent actually did in a session, or share a sanitized session map with others.
Who is agentplay not ideal for?+
Teams wanting centralized, real-time observability across many machines; agentplay reads local transcripts on one machine, and fidelity for non-Claude agents is limited to the main lane.
What are the best agentplay alternatives?+
The closest agentplay alternatives on ai.dosa.dev are Pieces for Developers, agenttrace, Vibe Kanban - all listed under Developer Productivity & Workflow.
Who makes agentplay?+
agentplay is developed by Entelligence AI. It is listed in the Developer Productivity & Workflow category on ai.dosa.dev.
Favorite this tool to revisit it later, or Zap it to contribute to the public vote count.