Skip to content
Get packs

Work Execution Command

Implements a plan or clear build request end-to-end and verifies it locally

Source Download skill
Repository
EveryInc/compound-engineering-plugin
Stars
25,850
Licence
MIT
Files
15 files

What it does

When you hand the agent a plan document, a spec, or a concrete build request, this skill makes it work through the tasks in order, make the code changes, and run local checks to confirm they work. It follows a set of bundled reference files that govern how it picks an execution approach, whether to split work across parallel helpers, how it commits changes, and when to stop and report a blocker. In standalone use it then hands off to a code review and shipping step; an outer workflow can instead use it to get only the implementation and verification result back.

Great for

  • Developers who already have a written plan or spec and want the agent to carry out the build
  • Teams running an orchestrator workflow that needs implementation and local verification without the delivery steps

Skip it if

  • Open-ended bug hunts, which the skill itself says should go to a debugging skill instead (the agent may do the wrong kind of work)
  • Repositories where you have not reviewed the changes: the skill edits files and makes git commits, so unreviewed runs can touch code you did not intend

Try asking

  • “Implement the plan in docs/plans/user-auth-plan.md”
  • “Build the CSV export feature described in this spec file, and verify it locally”
  • “Resume the run with id abc123 and report its status”
Step-by-step guide: how to use Work Execution CommandInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 15 original files, the MIT licence and an attribution note.

    ce-work.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip ce-work.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip ce-work.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Give it a specific plan path or a clear build request; vague prompts get sized and may be sent back for clarification
  • Review the git diff and commit history before pushing, since the skill makes incremental commits
  • Confirm which branch it will use; it is designed not to write on the default branch unless you direct it

What's inside

  • references/agents/figma-design-sync.md
  • references/agents/implementation-worker.md
  • references/cross-model-execution.md
  • references/execution-engines.md
  • references/execution-strategy.md
  • references/implementation-loop.md
  • references/implementation-result-schema.json
  • references/input-triage.md
  • references/multi-pr-delivery.md
  • references/non-code-execution.md
  • references/return-to-caller.md
  • references/review-findings-followup.md
  • references/shipping-workflow.md
  • references/tracker-defer.md
  • SKILL.md

Redistributed unmodified from EveryInc/compound-engineering-plugin under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.