groove-work-exec — Claude Code skill: run the plan
This skill takes a finished implementation plan and writes the code, tests and docs for it, logging progress in your task backend. Use it once planning is done.
npx skills add andreadellacorte/groove --skill groove-work-exec -a claude-codeWhat groove-work-exec does
groove-work-exec picks up where /groove-work-plan left off and works through that plan. The output is code, tests and artifacts complete enough to hand to /groove-work-review. Progress lands in the work task body as dated notes.
It reads tasks.backend from .groove/index.md to pick the backend. When tasks.backend != none, it creates a stage task through /groove-utilities-task-create titled YYYY-MM-DD, Work. Notes get appended to the task body as work moves along; nothing is overwritten. Marking the task complete is off the table here — that happens after review.
A prior plan is required. Without one the skill warns you and asks you to confirm scope before going further, so firing it straight after an idea won't get you far. Its allowed tools are Read, Write, Edit, Glob, Grep, Bash(git:*) and AskUserQuestion.
What it helps with
- execute a plan and get code ready for review
- track implementation progress inside the task body
- create a stage task titled with the date
- append progress notes without losing earlier ones
- catch the case where no plan exists yet
Example prompts for the agent
- execute the plan from groove-work-plan and keep progress notes in the task
- continue the implementation and append a dated note to the task body
- check tasks.backend in .groove/index.md and create today's stage task
How to install groove-work-exec
1Skills CLI (Vercel)
npx skills add andreadellacorte/groove --skill groove-work-exec -a claude-code2Manually: put the files in place
~/.claude/skills/groove-work-exec/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/groove-work-exec/
1Skills CLI (Vercel)
npx skills add andreadellacorte/groove --skill groove-work-exec -a codex2Manually: put the files in place
.agents/skills/groove-work-exec/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add andreadellacorte/groove --skill groove-work-exec -a cursor2Manually: put the files in place
.cursor/skills/groove-work-exec/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install andreadellacorte/groove groove-work-exec2Manually: put the files in place
.github/skills/groove-work-exec/SKILL.md1Skills CLI (Vercel)
npx skills add andreadellacorte/groove --skill groove-work-exec -a gemini-cli2Platform CLI
gemini skills install https://github.com/andreadellacorte/groove --path skills/groove-work-execManually: put the files in place
.devin/skills/groove-work-exec/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add andreadellacorte/groove --skill groove-work-exec -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about groove-work-exec
How do I install groove-work-exec in Claude?
Run npx skills add andreadellacorte/groove --skill groove-work-exec -a claude-code. The skill lands at ~/.claude/skills/groove-work-exec/SKILL.md.
What does groove-work-exec do?
It implements the plan produced by /groove-work-plan: writing code, tests and docs while tracking progress in the task backend. The result is ready for /groove-work-review.
What happens if there is no plan yet?
The skill warns you and asks you to confirm scope before proceeding. Running it before planning is not the intended flow.
Source text
SKILL.md · MIT
--- name: groove-work-exec description: "Execute the implementation plan, track progress in the task backend. Use after planning." license: MIT allowed-tools: Read Write Edit Glob Grep Bash(git:*) AskUserQuestion metadata: author: andreadellacorte --- # groove-work-exec ## Outcome Code, tests, and artifacts are written per the plan. The work task is created/updated in the backend with progress notes. Output is ready for review. ## Acceptance Criteria - Implementation matches the plan from `/groove-work-plan` - Work task body tracks progress with dated notes - Output (code, tests, docs) is complete enough to hand off to `/groove-work-review` - Stage task created/updated in backend: `YYYY-MM-DD, Work` ## Constraints - Read `tasks.backend` from `.groove/index.md` to determine backend - If no prior plan exists, warn the user and ask them to confirm scope before proceeding - Append progress notes to task body as work proceeds — do not overwrite - Do not mark task as completed during this stage — that happens after review - Create stage task via `/groove-utilities-task-create` if `tasks.backend != none` (title `YYYY-MM-DD, Work`)
© andreadellacorte · MIT · andreadellacorte/groove
