Omc-Omx_Harness

Clean AI-generated code slop with a regression-safe, deletion-first workflow and optional reviewer-only mode

Process-first advisor routing for Claude, Codex, or Gemini via `omc ask`, with artifact capture and no raw CLI assembly

Full autonomous execution from idea to working code

Cancel any active OMC mode (autopilot, ralph, ultrawork, ultraqa, swarm, ultrapilot, pipeline, team)

Given a target Android app — an APK path, a running emulator serial, or an installed package id — scope every screen and build a high-fidelity cloned RN/Expo app forked from app-template, filled with original mock content. Runs a 6-stage Acquire→Scope→Scaffold→Build→Parity→Deliver pipeline with 3 hard gates. Triggers: "clone app", "clone this APK", "clone <app> for me", "high-fidelity RN clone", "make a clone of this emulator app", /clone-app.

Configure notification integrations (Telegram, Discord, Slack) via natural language

Socratic deep interview with mathematical ambiguity gating before explicit execution approval

Deep codebase initialization with hierarchical AGENTS.md documentation

Deliver a generated document (md/HTML) as PDF or Word and reveal it in Finder. Use when the user says "give me a PDF", "export to Word/docx", "in pdf", "open in finder", asks for a Korean version of a doc, or a session ends with a document deliverable. Encodes the one known-good md→PDF pipeline on this Mac (Chrome headless; pandoc/xelatex fallback with pinned CJK font) so each session stops reinventing a broken one.

Convert Obsidian/Markdown to styled DOCX for Korean government proposals and research reports. Styles all tables with navy header, horizontal row alternation, content-proportional column widths, 10pt font, tight cell margins, vertical center alignment. Preserves embedded images. Hancom HWP compatible. Triggers on requests like "DOCX 테이블 스타일", "한글 보고서 표 꾸미기", "정부 제안서 포맷 DOCX", "export Obsidian to styled Word", "all tables styled", "proposal document formatting".

Invoke for Fable-grade reasoning on any task that carries silent or irreversible risk — code or non-code. Triggers: the explicit `/fable`; or "think like fable", "fable mode", "be maximally rigorous / careful", "what would fable do", "don't let this be subtly wrong", or a high-stakes, hard-to-reverse decision where you want the deepest possible reasoning. Installs Fable's silent/loud-rationed cognition as a session-wide stance and, gated by it, drives a generalized plan→consensus→orchestrate→sync loop — autonomous up to the truly-irreversible. Best under max-thinking on the session's top-tier model.

Launch a task through the act-bounded pipeline — crystallize (deep-interview/ralplan if vague) → isolated worktree → autonomous loop (ralph/autopilot/ultrawork) to completion → verify → propose a draft PR. Refuses under safe-mode. Use when the user types /go <task> or says "run this to completion autonomously".

Alias for /go (muscle-memory guard — the user types /goal in prompts and it used to silently do nothing). Use when the user types /goal, or writes "/goal" inline in a message as a directive. Immediately invoke the go skill with the same arguments.

Generates an extremely detailed LLM handoff report of the current session's work and opens a new cmux tab with claude pre-loaded to continue. Use when switching contexts, ending a session, or handing off work to a fresh LLM.

Configure HUD display options (layout, presets, display elements)

Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.

Extract a learned skill from the current conversation

Remind the user to rebuild OMC after editing TypeScript when running from a local fork. Triggered automatically by the AI whenever it notices it (or the user) just changed a src/**/*.ts file in an OMC dev install.

Configure popular MCP servers for enhanced agent capabilities

Diagnose and fix oh-my-claudecode installation issues

OMC agent catalog, available tools, team pipeline routing, commit protocol, and skills registry. Auto-loads when delegating to agents, using OMC tools, orchestrating teams, making commits, or invoking skills.

Install or refresh oh-my-claudecode for plugin, npm, and local-dev setups from the canonical setup flow

CLI-team runtime for claude, codex, or gemini workers in tmux panes when you need process-based parallel execution

Strategic planning with optional interview workflow

Run a report-only triage tick over a repo/task and write a prioritized DIGEST of items needing human review. Modes - `run` (the tick), `queue` (consume the review queue), `propose` (attended L2 propose-only acting - human-invoked, never pushes/deploys/merges, refuses unattended), `arm` (GATED - refuses until the containment spike passes). All fail-closed logic lives in scripts/review_digest.py; this skill only invokes it. The automated tick NEVER acts (no push/deploy/merge/outbound); `propose` is attended, propose-only, and still never pushes/deploys/merges. Human-invoked. Aliases: review-digest, flywheel.

Self-referential loop until task completion with configurable verification reviewer

Consensus planning entrypoint that auto-gates vague ralph/autopilot/team requests before execution

Generic release assistant — analyzes repo release rules, caches them in .omc/RELEASE_RULE.md, then guides the release

Use first for install/update routing — sends setup, doctor, or MCP requests to the correct OMC setup flow

Manage local skills - list, add, remove, search, edit, setup wizard

Turn a repeatable workflow from the current session into a reusable OMC skill draft

Explain complex concepts with analogies, visual metaphors, recursive diagnostics, and PhD-level depth using consulting frameworks

N coordinated agents on shared task list using Claude Code native teams

Evidence-driven tracing lane that orchestrates competing tracer hypotheses in Claude built-in team mode

Parallel execution engine for high-throughput task completion

Structured visual QA verdict for screenshot-to-reference comparisons

LLM Wiki — persistent markdown knowledge base that compounds across sessions (Karpathy model)