Skip to content

Dezycro CLI

The Dezycro CLI (dezycro) brings Dezycro's workbook, verification, and PRD/TRD authoring flows directly into your coding agent. Once installed, eight commands appear in your agent session, and the Companion background loop quietly guards the work your agent does for you against silent regressions.

Supported agents: Claude Code and Cursor — one install covers every supported agent on your machine, including Cursor's app and CLI.

What you get

Commands for the day-to-day Dezycro flow without leaving your editor. The flows are identical everywhere; only the trigger spelling differs per agent:

Command What it does
/dezycro:init Connect a repo to a Dezycro workspace + project
/dezycro:work <feature> Start or resume work on a feature; binds the session to the workbook
/dezycro:author --doc prd \| trd Guided PRD/TRD authoring with a built-in quality gate
/dezycro:status Compact workbook snapshot for the active feature
/dezycro:sync Sync the current chunk of work to the workbook
/dezycro:publish-spec Auto-detect + upload an OpenAPI spec; triggers verifier-binary generation
/dezycro:verify Pull the verifier binary, run it locally, upload results
/dezycro:import-features Reverse-engineer a codebase into Dezycro features with PRDs + TRDs
Skill What it does
/dezycro-init Connect a repo to a Dezycro workspace + project
/dezycro-work <feature> Start or resume work on a feature; binds the session to the workbook
/dezycro-author --doc prd \| trd Guided PRD/TRD authoring with a built-in quality gate
/dezycro-status Compact workbook snapshot for the active feature
/dezycro-sync Sync the current chunk of work to the workbook
/dezycro-publish-spec Auto-detect + upload an OpenAPI spec; triggers verifier-binary generation
/dezycro-verify Pull the verifier binary, run it locally, upload results
/dezycro-import-features Reverse-engineer a codebase into Dezycro features with PRDs + TRDs

Cursor skill names use a - where Claude Code uses :/dezycro-work is the same flow as /dezycro:work. Workspace-authored skills follow the same rule: /acme:db-migration in Claude Code is /acme-db-migration in Cursor.

Companion — background guardrails that work the same in every supported agent:

  • Block tasks from going to DONE / PUSHED without recent passing verifier evidence
  • Nudge you to re-publish your OpenAPI spec when you touch a controller
  • Sweep the workbook for decisions, issues, and assumptions worth recording — only when you ask (e.g. "save progress") or right after a verify
  • Surface coverage gaps before you push

Everything is opt-in per-trigger. Companion stays out of your way unless something material happened.

Next steps