momomuchu

Bootstrap a new project or catch up an agent on an existing one in minutes by reading vision/goals/state and scaffolding the minimum brain files.

Produce the structural plan — component boundaries, contracts, data flow — plus a lightweight decision record, from a sealed charter.

Formalize intent into a spec artifact (Gherkin scenarios, acceptance criteria, criticality tags) before any non-md file may change.

Declare and audit the per-project disciplines (cadence, branching, testing, architecture, methodology) that the pluggable StageSkillMap resolves against.

Implement the minimum code to satisfy the charter's RED tests, one logical change per commit.

Cited, multi-source research pass for H/C decisions or true blockers — converts findings into requirements, risks, and rejected alternatives.

Bridge blueprint to forge — break an approved charter+blueprint into atomic, sequenced or explicitly-parallel steps, cross-critiqued once before trust.

Batch every open question into ONE numbered list instead of serial one-at-a-time interrogation when no safe reversible default exists.

Zero-unverified-work execution harness for coding agents. Use when the user invokes /make-no-mistakes, says "make no mistakes", "don't get this wrong", "this must be correct", "prove it works", or hands over work where a wrong result is costly. Turns any task into a verified pipeline: freeze what "correct" means BEFORE acting, execute in budgeted loops against hard deterministic gates, have work certified by an independent verifier that is never the implementer, and report with evidence. Never claims done without proof; fails loudly (STUCK / GAMING-DETECTED) instead of pretending.

Adversarial verification panel for make-no-mistakes. Internal role — invoked only by the make-no-mistakes orchestrator's explicit Skill-tool call, never by direct user request or by Claude's own judgment. Certifies one diff against one frozen dod.lock.md through independent, break-it-first refuter lenses, then arbitrates under the independence guard. Runs forked, with no access to the conversation that produced the diff.

Internal creation-side specialist for /make-no-mistakes. Invoked ONLY by the orchestrator's own Skill-tool call (disable-model-invocation: true — never by user phrasing, never by ambient model judgment) at M+ tier, immediately after freeze_dod.sh succeeds and before any Implementer runs. Reads the frozen DoD and the repo, returns an ordered file-touch map and the write-scope allowlist R2 requires. Never writes code, never touches disk.

Internal creation-side specialist for /make-no-mistakes. Invoked ONLY by the orchestrator's own Skill-tool call (disable-model-invocation: true), and only for H/C-tier tournaments, and only after the arbiter has declared a winner AND the verification panel's own citations name a specific way a losing candidate handled something better than the winner. Grafts that improvement into the winning diff. Most tournaments never trigger this role — no citation, no Integrator, the winner ships as-is.

Internal creation-side specialist for /make-no-mistakes. Invoked ONLY by the orchestrator's own Skill-tool call (disable-model-invocation: true) at M+ tier, in parallel with mnm-planner, immediately after freeze_dod.sh succeeds. Authors falsifying tests for every acceptance criterion in the frozen DoD, structurally blind to any implementation — there is nothing to see yet, because this role runs before the Implementer exists. Mechanizes R2's "≥1 concrete example per acceptance criterion" requirement.

ISO-grade gated pipeline for agent-run academic/deep research. 8 stages (protocol → search → triage → graded reading → claim-level extraction → synthesis → verification → GRADE labeling) with fail-closed external gate scripts between stages. Use whenever an agent must produce a literature review, deep research report, or any synthesis whose citations must be real and whose claims must be traceable. Triggers: "literature review", "deep research", "systematic review", "research pipeline", "verify citations", "PRISMA", "GRADE".

Run a multi-persona simulated cognitive-walkthrough UX audit against a web app you own. Spawns >=3 isolated persona subagents that crawl a happy-path task list, log named friction instances tagged to a pluggable heuristic set (default: the Nielsen 10), converge findings across personas, and emit a gate-checked findings report with an honest validity envelope. Use when asked to audit UX friction, find usability problems, or CI-gate a UX regression on a local/staging target.

DFS graph crawler + full frontend audit. Playwright crawls the live app as a graph (pages=nodes, links=edges), DFS from entry, audits EVERY reachable node: multi-breakpoint screenshots (320/768/1440/1920px), WCAG 2.1 AA (axe-core), state verification (loading/error/empty/success), dark mode, interactive states, performance (LCP/CLS/FCP), visual regression, console errors, API validation, hydration. Auth support (storageState), CI mode (exit 1 on CRITICAL, text-only), max-pages cap, crawl failure recovery. Auto-detects 7 frameworks: Next.js, Vite/React, Vue/Nuxt, SvelteKit, Astro, Remix, Angular. Actions: test, crawl, audit, screenshot, verify, scan, regress, diff. Outputs: HTML graph report, topology map, per-node pass/fail, severity-rated issues.

Extract transcripts and key insights from YouTube videos. Handles single or multiple URLs. Uses yt-dlp for reliable transcript extraction, supports all languages. Actions: get transcript, summarize video, extract insights, analyze video. Triggers on: YouTube URL, 'insight video', 'resume video', 'transcript', 'what does this video say'.

Deep edge case discovery and bug hunting skill. Systematically finds bugs, crashes, security issues, and data corruption risks in any codebase. Uses 12 attack vectors: type coercion, boundary values, null/undefined, race conditions, injection, overflow, encoding, state machines, concurrency, file formats, API contracts, and UI edge cases. Actions: hunt, scan, audit, fuzz, stress, probe, test. Targets: API endpoints, parsers, forms, state machines, database operations, file uploads, auth flows, payment flows, email templates, PDF generation. Outputs: categorized bug reports with reproduction steps, severity ratings, and fix suggestions.