theft-kit

Use when a conversation is getting long, drifting, juggling multiple tasks, re-pasting the same files, or showing stale/confused context — and at the start of any new task. Triggers: long thread, context bloat, topic switch, repeated file uploads, 'why is it forgetting', degraded answers, finished one task and starting another.

Use when a user wants to deconstruct a complex problem, business challenge, product decision, strategy question, career crossroads, or stuck situation by stripping assumptions, identifying first principles, rebuilding options from foundational truths, and choosing one high-leverage action.

Audit role-based access control and permission drift across product surfaces.

Add a new feature to an existing @fabrk/* package following monorepo conventions

Build or rebuild the ADR index and dependency graph in AgentDB

Query AgentDB with semantic routing, hierarchical recall, causal graphs, and context synthesis

Use when the user wants to reduce AI coding spend without reducing shipping speed. Covers model routing, prompt caching, context discipline, loop profiling, and rollout plans for Cursor, Claude Code, Aider, and similar coding-agent stacks.

When the user wants to optimize content for AI search engines, get cited by LLMs, or appear in AI-generated answers. Also use when the user mentions 'AI SEO,' 'AEO,' 'GEO,' 'LLMO,' 'answer engine optimization,' 'generative engine optimization,' 'LLM optimization,' 'AI Overviews,' 'optimize for ChatGPT,' 'optimize for Perplexity,' 'AI citations,' 'AI visibility,' 'zero-click search,' 'how do I show up in AI answers,' 'LLM mentions,' or 'optimize for Claude/Gemini.' Use this whenever someone wants their content to be cited or surfaced by AI assistants and AI search engines. For traditional technical and on-page SEO audits, see seo-audit. For structured data implementation, see schema-markup.

MANDATORY rules - use design system tokens, existing components, and INDX Swiss patterns

Define API versioning rules, compatibility guarantees, and deprecation paths.

Use when building software with AI by separating Architect (planning) from Builder (implementation), with files as source of truth instead of chat history.

Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with elements, verify page state, diff before/after actions, take annotated screenshots, check responsive layouts, test forms and uploads, handle dialogs, and assert element states. ~100ms per command. Use when you need to test a feature, verify a deployment, dogfood a user flow, or file a bug with evidence. Use when asked to "open in browser", "test the site", "take a screenshot", or "dogfood this". (gstack)

Extract structured data from web pages using browser automation and DOM queries

Automated browser testing with Playwright -- navigate, interact, screenshot, and validate UI

Multi-pass quality audit that rates code across 5 dimensions (security, performance, code quality, UX/DX, competitive analysis). Use when someone says: run judges, be brutal, audit this, is this production ready. Defaults to NEEDS WORK. No inflated scores.

Display HTML canvas content on connected nodes via the OpenClaw canvas host and bridge.

Create communication plans for product/process changes across internal and external stakeholders.

When the user wants to reduce churn, build cancellation flows, set up save offers, recover failed payments, or implement retention strategies. Also use when the user mentions 'churn,' 'cancel flow,' 'offboarding,' 'save offer,' 'dunning,' 'failed payment recovery,' 'win-back,' 'retention,' 'exit survey,' 'pause subscription,' 'involuntary churn,' 'people keep canceling,' 'churn rate is too high,' 'how do I keep users,' or 'customers are leaving.' Use this whenever someone is losing subscribers or wants to build systems to prevent it. For post-cancel win-back email sequences, see email-sequence. For in-app upgrade paywalls, see paywall-upgrade-cro.

Analyze CI test failures with parallel log analysis and root cause identification

Reduce AI build errors by enforcing clean context discipline: curated inputs, minimal file payloads, and explicit update loops.

Reverse-engineer and rebuild mobile, desktop, or web apps as pixel-perfect Next.js clones from screenshots or a URL.

Review code for project standards, design system, security, and best practices

When the user wants to create competitor comparison or alternative pages for SEO and sales enablement. Also use when the user mentions 'alternative page,' 'vs page,' 'competitor comparison,' 'comparison page,' '[Product] vs [Product],' '[Product] alternative,' 'competitive landing pages,' 'how do we compare to X,' 'battle card,' or 'competitor teardown.' Use this for any content that positions your product against competitors. Covers four formats: singular alternative, plural alternatives, you vs competitor, and competitor vs competitor. For sales-specific competitor docs, see sales-enablement.

Run structured competitor teardown across product, pricing, UX, positioning, moat, and opportunity gaps.

Create React components using the project's existing design system primitives and tokens

When the user wants to edit, review, or improve existing marketing copy, or refresh outdated content. Also use when the user mentions 'edit this copy,' 'review my copy,' 'copy feedback,' 'proofread,' 'polish this,' 'make this better,' 'copy sweep,' 'tighten this up,' 'this reads awkwardly,' 'clean up this text,' 'too wordy,' 'sharpen the messaging,' 'refresh this content,' 'update this page,' 'this content is outdated,' or 'content audit.' Use this when the user already has copy and wants it improved or refreshed rather than rewritten from scratch. For writing new copy, see copywriting.

Forecast engineering and AI operational costs from usage, growth, and routing assumptions.

Track token and model costs by workflow, identify leaks, and propose routing optimizations.

Analyze token usage patterns and recommend cost optimizations with estimated savings

Generate a cost report showing token usage and USD costs by agent and model

Technical guide for creating a new Paperclip agent adapter. Use when building a new adapter package, adding support for a new AI coding tool (e.g. a new CLI agent, API-based agent, or custom process), or when modifying the adapter system. Covers the required interfaces, module structure, registration points, and conventions derived from the existing claude-local and codex-local adapters.

Scaffold a new Claude Code plugin with proper directory structure, plugin.json, skills, commands, and agents

Create dashboard pages and features following established patterns

Audit data quality across completeness, accuracy, consistency, and freshness.

Audit code for design system violations - grid spacing, tokens, existing components

Govern design system standards, drift control, and adoption across teams.

Discover and recommend ruflo plugins based on your workflow, installed MCP tools, and current task

Design product experiments with hypothesis, primary metric, guardrails, sample plan, and stop criteria.

Extract a component from fabrk-dev boilerplate into the framework monorepo

Reorder Foodora orders + track ETA/status with ordercli. Never confirm without explicit user approval. Triggers: order food, reorder, track ETA.