Initial public release.
A five-step interview that turns a rough task idea into a high-level Claude Code prompt — outcome, guardrails, and a self-verification method instead of a numbered procedure — then routes it to the right execution mode: single agent, dynamic workflow, local loop, or cloud routine.
It drafts only. It never creates or schedules a cron job, routine, or file on your behalf.
Install — Claude Code (recommended)
/plugin marketplace add promptmetrics/prompt-workflow-architecture
/plugin install promptmetrics-prompt-workflow@prompt-workflow-architecture
/reload-plugins
Also auto-triggers on prompt- and workflow-shaped questions, no slash command needed.
Install — Claude.ai / Cowork
Download prompt-workflow-architect.zip below, then upload it via Customize ▸ Skills. Requires Code execution and file creation enabled in Settings ▸ Capabilities.
Note: the skill description is 878 characters — under Anthropic's documented 1024-character limit, and kept long so its trigger phrases stay in the matcher. Some upload paths enforce a tighter ceiling, so the plugin path above is the verified route.
See the README and docs/how-it-works.md for the full walkthrough.