start-work (Plan Execution Orchestrator)

An orchestration skill that executes an approved `.omo/plans` work plan by delegating every unit to parallel subagents and closing checkboxes only against captured evidence.

AutomationAdvanced★ 67,531⑂ 5,507AI score 7/10Last updated: Aug 9, 2026

What it does

Selects a Prometheus-style plan under .omo/plans/, writes work state to .omo/boulder.json, and registers every phase and checkbox as todos up front.
The root agent never implements. All implementation, testing, QA and review work is spawned to subagents, with independent lanes dispatched in one parallel burst.
Each checkbox must pass five gates — plan reread, automated verification, real manual-QA artifact, adversarial QA (9 trigger-mapped classes), and cleanup receipts — with evidence appended line-by-line to .omo/start-work/ledger.jsonl.
A worker's DoneClaim only becomes done after an independent verifier returns a confirmed verdict; anything else loops back to the executor.
--worktree, --make-pr and --ship cover the delivery lifecycle: task-owned git worktrees, PR creation, CI/review watching, merge, and worktree teardown.

Who it's for

Operators of long-running agent pipelines that separate planning (ulw-plan) from execution.
Engineers running large refactors or multi-file features who want parallel subagents plus evidence-based completion.
Users of the oh-my-openagent (omo)/Codex-style harness with multi_agent spawn tools and a Stop-hook continuation.

Examples

$start-work — with no plan present, it bootstraps ulw-plan to produce an approved plan, then executes immediately.
$start-work auth-refactor --make-pr — creates a task-owned worktree, fans out independent lanes, and hands off a reviewer-ready PR.
$start-work billing --ship — opens the PR, watches CI and review gates, fixes from the worktree, merges, then removes the worktree.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/HEAD/packages/shared-skills/skills/start-work/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)

Download the ZIP with the button below.
In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.

↓ Download ZIP

Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the packages/shared-skills/skills/start-work folder from the GitHub repo code-yeongyu/oh-my-openagent into my ~/.claude/skills/start-work/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/code-yeongyu/oh-my-openagent.git /tmp/oh-my-openagent && mkdir -p ~/.claude/skills && cp -r /tmp/oh-my-openagent/packages/shared-skills/skills/start-work ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

Clone the repo: git clone https://github.com/code-yeongyu/oh-my-openagent.git /tmp/oh-my-openagent
Create the skills folder: mkdir -p ~/.claude/skills
Copy the skill: cp -r /tmp/oh-my-openagent/packages/shared-skills/skills/start-work ~/.claude/skills/
Also copy the companion planning skill (e.g. ulw-plan from the same skills/ directory) so the no-plan bootstrap works.
In your project, create .omo/plans/ and add a plan .md file containing - [ ] checkboxes.
Restart Claude Code, then say "start work" or "execute plan" to trigger it.
Note: without subagent spawn tooling the parallel delegation rules won't fully apply — test with a small plan first.

View source on GitHub ↗License: NOASSERTION