oh-my-workbuddy

Run a regression-tests-first, smell-by-smell deslop/refactor workflow that preserves behavior and raises signal quality. Use when the user asks to "cleanup", "refactor", "deslop", or when follow-up code has duplicates, dead code, weak boundaries, or fallback-like slop.

Run read-only deep repository analysis and return a ranked synthesis with explicit confidence, concrete file references, and clear evidence-vs-inference boundaries. Use when a user says "analyze", "investigate", "why does", "what's causing", or needs grounded cross-file explanation before any changes are proposed.

Deprecated compatibility shim for Claude advisor requests. Use `ask` with the claude backend instead.

Deprecated compatibility shim for Gemini advisor requests. Use `ask` with the gemini backend instead.

Ask a locally installed external advisor CLI (Claude or Gemini) for focused questions, reviews, brainstorming, or second opinions, and capture a reusable artifact. Use when the user wants a Claude/Gemini second opinion, "ask claude", or "ask gemini".

Strict autonomous delivery loop that runs deep-interview, ralplan, ultragoal, code-review, and ultraqa in order with automatic re-planning when gates are not clean. Use on "autopilot", "autonomous", "build me", "create me", "make me", "full auto", "handle it all", or "I want a/an…".

Durable professor-critic research workflow bound to a tracked goal, with validation-gated completion. Use when a research mission needs ongoing goal-focused management plus professor/critic-style validation.

Stateful validator-gated research loop that keeps nudging until explicit validation evidence exists. Use when a research output is a bounded deliverable that must pass a script or professor-critic validator, not for ordinary pre-planning docs lookup.

Bounded best-practice research using official/upstream evidence first; produces a cited, reusable recommendation and a handoff to planning/execution

Deprecated shim — build failures are now handled inline by the active execution/debugging workflow; do not route here

Abort the current WorkBuddy workflow — stop background agents/processes, clear in-progress tasks, and hand back control to the user

Run a comprehensive, severity-rated code review for quality, security, and maintainability, with a deterministic merge-readiness verdict

Configure session notifications (Discord/Telegram/Slack webhooks or custom commands) by writing a WorkBuddy-native notifications config file

Socratic deep interview with mathematical ambiguity gating that turns vague ideas into execution-ready specifications before planning or implementation

Deprecated shim — use the `analyze` skill for deep repository analysis or repo inspection tools (Read/Grep/Glob/Bash) for fast repo-local lookup

Canonical repo-local DESIGN.md workflow for product, UI/UX, and frontend decision source of truth

Diagnose and fix oh-my-workbuddy skill installation issues — missing or duplicate skill roots, stale package versions, broken AGENTS.md/memory. Trigger "doctor", "fix skill install", "why is my skill missing", "diagnose skills".

DEPRECATED shim — cost-optimized execution modifier. Do not route here; use the `ultrawork` skill for maintained high-throughput parallel execution instead.

DEPRECATED compatibility shim for frontend UI/UX work. Do not route here; use the `design` skill for design/UX context or `visual-ralph` for measured visual implementation.

Git expert for atomic commits, interactive rebasing, branch management, and history cleanup. Trigger "git rebase", "atomic commits", "clean git history", "squash commits", "branch strategy".

DEPRECATED shim. Use the `omx-setup` skill for install/setup and the `doctor` skill for troubleshooting instead.

Inspect live progress of long-running or multi-step work. WorkBuddy has a built-in status display; this skill explains how to read progress via the task list and memory. Trigger "show progress", "what's the status", "hud", "where are we".

DEPRECATED shim. Use the WorkBuddy memory surface `.workbuddy/memory/YYYY-MM-DD.md` directly when you need to persist notes.

Install or refresh the oh-my-workbuddy skill package into user or project scope. Trigger "omx setup", "install skills", "set up oh-my-workbuddy", "refresh skills".

Run an evaluator-gated performance optimization loop with durable WorkBuddy artifacts and safe handoffs. Use when the user wants measurable performance/optimization work (latency, throughput, memory) driven by a pass/fail benchmark rather than a one-off review. Triggers: 'optimize performance', 'reduce latency', 'performance goal', 'benchmark-driven optimization'.

Configurable orchestrator that sequences planning/execution/review stages as ordered WorkBuddy skill invocations with a task-list state and resume support. Use to run a repeatable multi-stage workflow (interview → plan → execute → review → QA). Triggers: 'run the pipeline', 'autopilot pipeline', 'sequence stages', 'stage orchestrator'.

Strategic planning with optional interview, consensus (Planner→Architect→Critic), and review modes. Use to scope vague ideas, produce quality-gated work plans, or review an existing plan before implementation. Triggers: 'plan this', 'let's plan', 'ralplan', 'review this plan', 'plan --consensus', 'plan --review'.

Clean-room interview-driven planner with three voices (Metis clarifies, Momus challenges, Oracle synthesizes), then hands off to ultragoal/team. Use for rigorous pre-execution planning when ambiguity is risky. Triggers: 'prometheus strict', 'strict plan', 'interview-driven plan', 'rigorous planning before execution'.

Deprecated. Do not invoke. Use ralph directly after PRD/test-spec planning is complete.

Self-referential persistence loop that keeps working on a task until it is fully complete and architect-verified, with parallel delegation, automatic retry, and a mandatory verification gate. Use when guaranteed completion matters. Triggers: 'ralph', 'don't stop', 'must complete', 'finish this', 'keep going until done'.

Alias for plan --consensus. Triggers iterative Planner→Architect→Critic consensus planning with RALPLAN-DR structured deliberation. Use when you want multi-perspective sign-off on a plan before execution. Triggers: 'ralplan', 'consensus plan', 'plan --consensus'.

Deprecated. Do not invoke. Use code-review directly for implementation review, or plan --review only when reviewing an existing planning artifact.

Deprecated standalone security review skill. Use `code-review` when security concerns are in scope.

Manage local WorkBuddy skills - list, add, remove, search, edit, info, sync, setup. Use for skill authoring and inventory.

Deprecated compatibility shim for team execution. Use `team` (Agent-tool coordination) instead.

Deprecated TDD shim. Keep test-first discipline inside the active implementation workflow (folded into `plan`).

Coordinate N parallel agents on a shared task list using the Agent tool and TeamCreate. Use for multi-lane execution, fan-out, shared delivery+verification, and long-running lanes tracked across a session.

Deprecated trace/runtime-inspection shim. Use normal repo inspection (Read, Grep, Glob, Bash) instead.

Create and execute durable repo-native multi-goal plans. Break a brief into tracked goals, drive them to verified completion, and steer as findings change.

Adversarial dynamic end-to-end QA - generate hostile scenarios, test, verify, fix, report, and clean up. Use for robust behavior verification.