$ loading_
Coordinate multi-agent coding with swappable roles, persistent memory, and replay.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "threadwork" yet — see the docs or source repo.
Use threadwork with Claude Code to configure three roles: architect, backend developer, and code reviewer. Have them collaborate on a Node.js API with user login and authorization, and save discussion and decision records at each step.
A development result containing role-based contributions, code implementation, review feedback, and a traceable process log.
In an existing threadwork session, hot-swap in a "database performance specialist" role to analyze current SQLite query bottlenecks, propose indexing and SQL optimizations, and continue with preserved context memory.
Performance diagnostics, optimization recommendations, and updated collaboration records based on preserved context.
Use threadwork's step-level replay to review the full collaboration process from the last production bug fix, then summarize key decisions, failed attempts, and the final solution for a retrospective.
A replayable incident timeline, key decision summary, and final fix explanation for retrospective review.
Coordinate multiple Claude Code agents for collaborative development and automated code review.
Set explicit temporary rules for AI coding threads without changing project policy.
Give AI coding agents verified continuity memory across sessions and tasks.
Self-host and orchestrate multi-agent workflows with memory, scheduling, and multi-LLM support.
Production-ready Claude Code workflows powered by specialized AI agents.
Query multiple AI models for code reviews, debates, and diverse perspectives.