hive

Ultra-terse brand+visual design partner for identity systems, UI mockups, and conversion-minded visuals.

Ultra-terse Playwright guardian focused on modular, stable, coverage-rich E2E tests.

Ultra-terse n8n automation expert for workflows, integrations, and reliability.

Ultra-terse Vercel deployment + performance engineer for CI/CD, edge, and monitoring.

Ultra-terse backend lead for APIs, data models, auth, and performance.

Ultra-terse SEO specialist. Illuminates content, quantifies visibility, delivers data-backed optimizations.

Ultra-terse Next.js/React implementer focused on UI quality, performance, and accessibility.

Ultra-terse web research specialist. Relentless investigator that uses Perplexity to find, cross-reference, and distill real-time information for informed decision-making. Use when needing market research, competitive analysis, technology evaluation, fact-checking, or any decision requiring current, verified web data.

Ultra-terse Next.js/React implementer focused on UI quality, performance, and accessibility.

Ultra-terse brand strategist and art director. Defines identity, design systems, and visual direction — Stitch executes.

Ultra-terse testing guardian. Vitest for unit/integration, Playwright for E2E. Coverage-rich, stable, modular.

Ultra-terse n8n automation expert for workflows, integrations, and reliability.

Ultra-terse Vercel deployment + performance engineer for CI/CD, edge, and monitoring.

Ultra-terse backend lead for APIs, data models, auth, and performance.

Ultra-terse SEO specialist. Illuminates content, quantifies visibility, delivers data-backed optimizations.

Ultra-terse web research specialist. Relentless investigator that uses Perplexity to find, cross-reference, and distill real-time information for informed decision-making. Use when needing market research, competitive analysis, technology evaluation, fact-checking, or any decision requiring current, verified web data.

Ultra-terse marketing strategist. Content creation, social media management, campaign planning, audience research, and brand positioning for B2B LATAM market. Use when needing content calendars, social posts, email campaigns, competitive positioning, audience analysis, or marketing strategy.

Es un registry de Vercel (open source via vercel-labs/skills) donde se instalan skills con npx skills add <repo>. Básicamente son archivos markdown que inyectan conocimiento procedimental a agentes AI (Claude Code, Cursor, Copilot, etc). Son exactamente lo mismo que nuestros skills en hive/skills/, pero con distribución pública.

**Purpose:** Guide Mentat's decisions on when and how to invoke Forge (GPT-5) or Mycelium (Gemini 2.5) for technical review, adversarial feedback, and system-wide consistency checks.

Manage issues across Linear workspace. Use when user wants to create, update, move, or query issues. Routes to appropriate command based on workflow state.

3D modeling and rendering via Blender MCP. Use when creating 3D models, architectural visualization, product renders, or any 3D scene. Requires Blender open with MCP addon connected.

Write blog posts for skywalking.dev in Mentat's voice. Research-backed, concise, opinionated.

Capture ideas quickly into Linear backlog with minimal friction. Use when user has a quick idea, bug report, or thought that needs to be captured without full discovery process. Creates issue in Backlog state with title and brief context.

B2B cold email outreach for a dev studio targeting CTOs, engineering managers, and technical founders. Use when writing first-touch outreach emails, follow-up sequences, LinkedIn connection messages, or WhatsApp follow-ups for business development. Triggers on "cold email", "outreach email", "write to a CTO", "follow-up sequence", "BD email", "prospecting message".

Competitive positioning, "alternatives to X" content, and battle cards. Use when asked to compare against competitors, create an alternatives page, build a sales battle card, or find positioning gaps. Triggers on "how do we compare to X", "alternatives to X", "battle card", "positioning against agencies", "why choose us over X", "competitor analysis".

Content planning and editorial strategy for a technical B2B audience. Use when building content calendars, selecting blog topics, planning case studies, mapping content pillars, or creating a publishing roadmap. Triggers on "content calendar", "blog strategy", "what should we write about", "content pillars", "editorial calendar", "content plan", "SEO topics".

Conversion copywriting for dev tools, SaaS landing pages, and technical services. Use when writing homepage copy, feature pages, pricing pages, case studies, hero sections, CTAs, or any web copy targeting developers, CTOs, or technical buyers. Triggers on "write copy for", "landing page copy", "hero section", "case study copy", "pricing page", "value proposition", bilingual copy requests (ES/EN) for technical products.

Guide Mentat's delegation decisions to specialist agents. Use when routing tasks to sub-agents (Kokoro, Pixel, Aurora, Flux, Hermes, Centinela, Lumen) based on Linear sub-issues with [Agent] title prefix.

Analyze Stitch projects and synthesize a semantic design system into DESIGN.md files

Orchestrate work on Linear issues by delegating sub-issues to specialist agents. Use when starting work on an issue with [Agent] sub-issues. Mentat monitors progress, resolves doubts, and validates acceptance criteria.

Design email sequences for lead nurturing, onboarding, and re-engagement. Use when asked to create a welcome sequence, onboarding drip, re-engagement campaign, or any multi-email nurture flow. Triggers on "email sequence", "welcome emails", "onboarding drip", "nurture campaign", "follow-up sequence", "re-engagement emails", "email automation".

Use this skill when building applications with Gemini models, Gemini API, working with multimodal content (text, images, audio, video), implementing function calling, using structured outputs, or needing current model specifications. Covers SDK usage (google-genai for Python, @google/genai for JavaScript/TypeScript, com.google.genai:google-genai for Java, google.golang.org/genai for Go), model selection, and API capabilities.

Generate speech via OpenAI TTS — Mentat's voice. Use when user wants to hear a response spoken, generate audio briefings, or create voice content. Supports ES-LATAM and EN-UK.

Generate images via Gemini API (Imagen 4.0 or Gemini Flash). Use when user needs to create, edit, or transform images. Supports text-to-image and image editing with input reference.

GitHub CLI operations for commit, push, PR creation, reviews, and branch housekeeping. Use for git workflows, PR management, branch cleanup, review comments, and GitHub API interactions via `gh` CLI.

Read and search Gmail inbox, manage drafts, organize with labels, and create filters. Use when user needs to check email, compose drafts, or organize messages.

Read Google Calendar events, check availability, and search events. Use when user needs schedule info, today's events, or free/busy status.

Read, create, and manage Google Docs with Markdown conversion. Use when user needs to read, write, search, or export Google Docs.

Manage files and folders in Google Drive - list, search, download, upload, share, and organize. Use when user needs Drive file operations.

Access any Google Workspace API via gws CLI or MCP tools. Use for Sheets, Tasks, Chat, Slides, People, Forms, Meet, and as fallback for Drive/Gmail/Calendar/Docs when existing handlers lack a feature. Use existing Python handlers first for Gmail (send-blocking), Drive (domain allowlist), Calendar, Docs.

Manage Instagram Business account - publish posts/reels/stories/carousels, view insights, manage comments, search hashtags. Use when user needs to post content, check performance, or moderate comments on Instagram.

Systematic debugging — prevents jumping to fixes without root cause analysis. Use when debugging errors, analyzing failures, or investigating unexpected behavior. Forces evidence-based investigation before any code changes.

Research and update the LLM API Landscape document with current pricing, providers, free tiers, and recommendations.

Manage Meta Ads campaigns, ad sets, insights, audiences, and creatives. Use when user needs to create/monitor ad campaigns, check performance, or manage audiences on Meta (Facebook/Instagram).

Conversion rate optimization for web pages. Use when asked to audit, improve, or critique a landing page, hero section, pricing page, or any product page for a SaaS or dev tool. Triggers on "optimize this page", "why isn't this converting", "improve the CTA", "audit our landing page", "hero section feedback", "page isn't converting".

Search the web and query LLMs with real-time data via Perplexity API. Use when user needs web research, current information, grounded answers with citations, or wants to search using Perplexity's Sonar models. Triggers on web search, research queries, "ask perplexity", current events, or real-time data needs.

Blameless incident postmortem — gathers data from Slack, Linear, and user input to produce structured incident documentation. Three modes: extract from raw dumps, start from scratch, or refine existing drafts.

Review an EXISTING open PR and sync Linear state. Use for manual review of another agent's or teammate's PR, or post-merge Linear transitions. For local pre-push self-review during /ship_it, use /review-diff instead — /pr-review operates on GitHub PRs and is the post-push layer.

Pricing strategy for SaaS products and dev studio services. Use when asked to design pricing tiers, set rates, evaluate freemium vs paid, create packaging, or adapt pricing for LATAM markets. Triggers on "how should we price this", "pricing tiers", "freemium or paid", "what should we charge", "pricing page", "LATAM pricing", "consulting rates".

**Propósito:** Procesar archivos de audio (conversión + transcripción) de forma batch o individual

Watch: extract transcripts from YouTube or local video files, then analyze content for topics, insights, and connections. Replaces extract_transcript + video-analysis as a single sense.

Commit, push, and open a PR in one step. Use when user says "push it", wants to commit + push + PR, or is done coding and ready for review.

Converts Stitch designs into modular Vite and React components using system-level networking and AST-based validation.

Refine shaped issues into technical breakdown with agent sub-issues. Use when an issue has been shaped and needs technical planning - identifying agents, creating sub-issues with specific tasks, and adding technical specs (API, DB, components).

**Purpose:** Structured alignment between 2-5 agents → concrete decisions, not transcripts.

Local pre-push code review by Claude. Analyzes git diff vs base branch and returns structured findings (CRITICAL/HIGH/MEDIUM/LOW) without touching GitHub or Linear. Use before pushing to catch issues in-session, fix them, and ship a clean PR. Different from /pr-review which is the post-push CI/Linear sync layer.

Expert guidance for integrating and building applications with shadcn/ui components, including component discovery, installation, customization, and best practices.

Shape bugs, features, and improvements into well-defined Linear issues. Use when user wants to create a new ticket, report a bug, request a feature, or propose an improvement. Provides guided discovery via AskUserQuestion and creates issues directly in Linear with proper structure and sub-issues for agents.

Full shipping pipeline — lefthook + local review + fix + simplify + push + CI review + merge + deploy + verify + close Linear + report. Use when user says "ship it" and wants code to reach production verified. Handles everything from uncommitted changes to live deployment, with review happening PRE-push so PRs nacen verdes.

Interact with Slack via Python script. Read threads, post messages, upload files. Use when user asks for Slack content, mentions thread URLs, or needs to send messages.

Social media content for LinkedIn (primary) and X/Twitter targeting developers and technical buyers. Use when writing LinkedIn posts, X threads, company page content, founder posts, or engagement copy. Triggers on "LinkedIn post", "write a tweet", "X thread", "social post", "founder content", "what to post on LinkedIn".

Unified entry point for Stitch design work. Handles prompt enhancement (UI/UX keywords, atmosphere), design system synthesis (.stitch/DESIGN.md), and high-fidelity screen generation/editing via Stitch MCP.

Query Supabase database and manage projects via MCP. Use when user needs database queries, migrations, auth operations, or logs. Triggers on SQL, table names, auth, migrations.

Generate technical construction drawings (floor plans, elevations, sections) as ISO-standard SVG/HTML sheets. Use when creating architectural plans, construction documentation, or any dimensioned technical drawing for EterHabitats or other construction projects.

Execute and debug E2E tests collaboratively with real-time log monitoring. Use when running Playwright or similar E2E tests where both Claude and the user need to observe execution simultaneously - Claude monitors logs while user watches the browser. Supports headed/UI mode execution, log capture, real-time monitoring, screenshot analysis, and collaborative debugging workflows.

Deploy, manage environment variables, view logs, and configure cron jobs with Vercel CLI. Use when deploying to Vercel, managing env vars, viewing logs, or configuring cron in vercel.json.

Send WhatsApp Business messages via Cloud API v22 - text, templates, media, interactive buttons/lists, flows, location, contacts, reactions. Use when user needs to send messages, check templates, or manage WhatsApp Business.

End-of-session summary and conversation rename. Use when user says "/wrap", "wrap", "resumen de sesion", or signals they're done working. Generates a structured summary of what was accomplished and renames the conversation.

A Model Context Protocol server for interacting with Chrome DevTools, enabling browser automation, debugging, and performance analysis capabilities.

Google Stitch — design-to-code, UI generation, React components

Trivago — hotel search and price comparison