Ultrawork — Maximum-Precision Work Mode
Typing 'ultrawork' or 'ulw' flips the agent into a strict failing-first, evidence-driven engineering mode with mandatory cleanup receipts.
What it does
A binding directive that activates whenever a prompt contains ultrawork or ulw. It overrides the agent's default "looks fine, done!" habit and enforces:
- Tier triage (LIGHT/HEAVY): classified once at bootstrap; auth, DB migrations, external integrations, concurrency, or cross-domain refactors force HEAVY, and downgrading mid-task is forbidden.
- A registered, binding goal: deliverable, success criteria, literal PASS/FAIL observables, evidence artifacts, and an explicit "when to stop" line written upfront.
- RED → GREEN → SURFACE → CLEAN loop: capture a failing proof before any production code, then prove the behavior on the real surface (
curl -i, a real browser, an xterm.js-rendered TUI screenshot), then tear down every spawned server, tmux session, container, and temp file with a written cleanup receipt. - A durable notepad: an append-only
mktempmarkdown log of plan, findings, commands, RED/GREEN captures, and artifact paths that survives context compaction. - Parallel & delegation discipline: batch independent lookups into one wave, fan out subagents only when write scopes are disjoint.
- Hard bans: no skipping/
.only/deleting tests to green a suite, no lint suppression, no claiming done from inference.
Who it's for
- Developers tired of agents reporting success on work that doesn't actually run.
- Leads who want TDD and captured evidence enforced rather than suggested.
- oh-my-openagent (omo-senpi) users, or anyone mining a well-designed agent workflow for their own rules.
Caveat: it assumes omo-senpi tooling (
create_goal,task,team_create,eval, senpitodo). On stock Claude Code only the fallback paths (notepad, checklist) apply.
Examples
ulw add email validation to the login API→ failing test captured RED first, then GREEN plus acurl -i400 response body as surface evidence.ultrawork refactor the payment webhook integration→ classified HEAVY: 3+ criteria (happy, edge, regression), characterization tests pinned first, reviewer loop when a plan file exists.ulw fix the broken colors in the TUI→ visual proof rendered through a real xterm.js web terminal screenshot instead oftmux capture-pane, with a session-kill receipt logged afterward.
· · · 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/omo-senpi/skills/ultrawork/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.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on Claude Skill Mart. Copy the packages/omo-senpi/skills/ultrawork folder from the GitHub repo code-yeongyu/oh-my-openagent into my ~/.claude/skills/code-yeongyu-ultrawork/. 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 && mkdir -p ~/.claude/skills && cp -r oh-my-openagent/packages/omo-senpi/skills/ultrawork ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal.
- Clone the repo:
git clone https://github.com/code-yeongyu/oh-my-openagent.git - Create the skills folder:
mkdir -p ~/.claude/skills - Copy the skill:
cp -r oh-my-openagent/packages/omo-senpi/skills/ultrawork ~/.claude/skills/ - Verify that
~/.claude/skills/ultrawork/SKILL.mdexists. - Restart Claude Code, then prefix a request with
ulworultrawork, e.g.ulw fix the login bug. - You'll know it fired when the first line of the reply is
ULTRAWORK MODE ENABLED!. - (Optional) If your environment lacks omo-senpi tools, edit the
create_goal/task/team_createsections of SKILL.md to match the tools you actually have.