catcatcatstudio

eat

Extract knowledge, frameworks, and methodologies from a URL, file, video, article, or podcast. Use when user says '/eat', 'eat this', 'eat from', shares a URL/file for insights, or wants to learn from a video/article without reading the whole thing. NOT for summarization or news digests. Requires yt-dlp, whisper or GROQ_API_KEY, ffmpeg.

fortify

Testing infrastructure, coverage, and verified test quality for any project. Detects stack, installs testing ecosystem, audits untested critical paths, writes thorough tests weighted toward error paths, runs them, verifies via mutation testing. Use when user says '/fortify', 'add tests', 'set up testing', 'I need tests', 'test this', or wants to audit coverage. NOT for: running existing tests directly or improving prompt quality.

memento

Context handoff for fresh sessions. Saves unsaved session knowledge to notebook, then produces a short orientation block to paste into a new chat. Use when user says '/memento', 'hand off context', 'prepare for fresh chat', 'save and brief', or before clearing context. NOT for: recovering context (use /notebook recover) or saving a single note.

scorched-earth

Obsessively thorough analysis. Three modes: build (find every issue in a codebase), intake (extract everything from a URL or reference), research (exhaustive investigation of a topic). Use when surface-level isn't enough and you need certainty nothing was missed. Triggers: '/scorched-earth', 'go scorched earth', 'find everything', 'leave no stone unturned', 'I need to be thorough'.

vet

Thorough codebase investigation that finds and fixes real problems. Maps architecture, scans every layer (structural, correctness, security, reliability, performance, hygiene), prioritized report with evidence, fix plan, executes collaboratively. A senior engineer doing a full review with authority to fix. Use when user says '/vet', 'vet this project', 'check this codebase', 'find all the problems', 'what's broken', 'audit this code', or wants to clean up AI-generated code. NOT for: UI review (/eye), test writing (/fortify), pre-build research (/recon).

xray

X/Twitter content intelligence with a social graph. Scout reply opportunities, pulse- check topic lanes, track competitors, mirror your accounts, prospect clients. Graph- first — built around accounts and topics YOU care about, not generic search. Use when user says 'xray', 'x research', 'search x', 'scout', 'pulse', 'track', 'mirror', 'prospect', '/xray', or needs real-time X discourse / engagement opportunities. NOT for: posting tweets or analytics dashboards.

adversary

Adversarial pressure-test of a decision, plan, or piece of work. Articulates the case FOR the current approach, then spawns a subagent to steel-man the best alternative, returns a synthesis with mandatory verdict. Invoke ONLY via /adversary — do NOT trigger on 'what do you think' or 'any concerns'.

architect

Staged build methodology for software projects: technical spike → design notes → build plan → stage-by-stage execution with professional-grade testing. Use when user says '/architect', 'new project', 'start a project', 'build this from scratch', wants to plan/build with structured methodology, or resumes an in-progress staged build. NOT for: quick scripts or one-off fixes.

elevate

Shifts the model from compliant executor to critical expert advisor. Identifies the domain, adopts top-tier expert perspective, understands the PURPOSE, returns ranked proposals to elevate the work. Works across UI/UX, copy, architecture, planning, code, strategy, branding. Invoke ONLY via /elevate — do NOT trigger on 'make this better' or 'what am I missing'.

eye

Design judgment via design-director taste — reads design intent, assesses whether execution serves it, returns opinionated verdict with clear direction. Scales from one element to a whole app. Always judgment, never fixing. Invoke ONLY via /eye — do NOT trigger on 'review the design' or 'how does it look'.

liquid-cat-physics

Autonomous deep-work agent loop for driving projects forward. Each invocation: reads project state, applies critical expert thinking, executes one focused unit of work, persists to notebook + PROJECT_STATE.md. Sets up /loop cron on first run (default 10m). Three-tier confidence gate (GREEN/YELLOW/RED) blocks action when human input is needed. Invoke ONLY via /liquid-cat-physics or 'lcp' — do NOT trigger on 'autonomous build' or 'deep work loop'.

notebook

Project notes system to prevent AI context loss and reasoning loops. Init notes, save mid-flow, recover context, migrate messy notes, resolve lessons. Subcommands: /notebook, /notebook save, /notebook recover, /notebook migrate.

prodev

Pro mode — engineering standard enforcement. Forces senior-engineer behavior: root-cause discipline, layer awareness, current-gen API verification, blast-radius thinking, caller-first API design, no stubs, no sycophantic agreement, no unverified assumptions. Invoke via /prodev or 'prodev' — always run via Skill tool, never skip or summarize.

recon

Pre-build decision intelligence. Produces opinionated decision brief — the senior- consultant week-1 output before scope, stack, or approach are locked. Surfaces traps and edges, calibrated to project scope (prototype vs production vs enterprise). Use when user says '/recon', 'what to watch out for building X', 'where's the edge', 'common mistakes with X', starts a new project, or hits domain uncertainty mid-build. NOT for: debugging, single-API research, URL extraction (/eat), or post-build optimization (/fortify).