AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

ulw-plan (Prometheus Planning Consultant)

Turns Claude into an explore-first planning consultant that grounds in your codebase, asks only the decisions you must own, and — only after your explicit okay — writes one decision-complete work plan.

Dev & CodingAdvanced★ 67,531⑂ 5,507AI score 9/10Last updated: Aug 9, 2026

What it does

  • Locks the agent into the Prometheus planner persona: no product-code edits, no implementer subagents, until you explicitly approve — and approval only authorizes writing the plan, not executing it.
  • Runs read-only exploration first (Read/Grep/Glob, CodeGraph when available, parallel research subagents) and brings you only genuine owner-decisions.
  • Routes intent to CLEAR (you know the outcome → asks only surviving forks) or UNCLEAR (fuzzy brief → researches, announces best-practice defaults, and auto-runs high-accuracy review).
  • Uses scripts/scaffold-plan.mjs to create a compaction-safe draft at .omo/drafts/<slug>.md, then .omo/plans/<slug>.md after approval — so a session can resume from durable state.
  • Enforces a strict plan grammar: column-zero task rows - [ ] N. title and verifier rows - [ ] F1. title, each with references, acceptance criteria, agent-executed QA, and commit boundaries.

Who it's for

  • Developers who want the AI to stop coding first and plan first on large refactors or features.
  • Teams handing work off across sessions or agents, using the plan file as the handoff contract.
  • Anyone who submits vague briefs and wants defensible defaults chosen for them instead of an interrogation.
  • Prerequisite: Node or Bun, a writable .omo/ directory, and an agent stack with spawn/wait tools (multi_agent_v1/v2); Codex CLI is assumed for the second high-accuracy review.

Examples

  1. ulw-plan add a 5/min-per-IP rate limit to /login → routed CLEAR; it inspects existing middleware, asks only the real fork (Redis vs in-memory), then writes the plan after your okay.
  2. make auth better → routed UNCLEAR; no interview, it researches, announces best-practice defaults, and runs the dual high-accuracy review automatically.
  3. plan this DB migration, ultra high accuracy → sets review_required: true, requires momus plus an independent review before handoff; execution starts separately via $start-work.

· · · 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-codex/plugin/skills/ulw-plan/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)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. 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/omo-codex/plugin/skills/ulw-plan folder from the GitHub repo code-yeongyu/oh-my-openagent into my ~/.claude/skills/code-yeongyu-ulw-plan/.
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-codex/plugin/skills/ulw-plan ~/.claude/skills/

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

  1. Open a terminal in a folder where you keep source checkouts.
  2. Clone the repo: git clone https://github.com/code-yeongyu/oh-my-openagent.git
  3. Create the skills folder: mkdir -p ~/.claude/skills
  4. Copy the skill: cp -r oh-my-openagent/packages/omo-codex/plugin/skills/ulw-plan ~/.claude/skills/
  5. Verify ~/.claude/skills/ulw-plan contains SKILL.md, the references/ folder, and scripts/scaffold-plan.mjs — the skill depends on all three.
  6. Confirm Node.js is available with node -v (Bun also works).
  7. Restart Claude Code and try: ulw-plan add rate limiting to /login. You should see ULW-PLAN MODE ENABLED! as the first line.
  8. Review the approval brief before saying okay; plan artifacts land in your project's .omo/ directory, and execution is a separate step ($start-work).
View source on GitHub ↗License: NOASSERTION