sanchocmo-openclaw

Sistema de personalización por niveles para outbound B2B, destilado de ColdIQ: cuándo personalizar en masa vs en profundidad, los 6 buckets de research rankeados, hooks fuertes vs ligeros, calibración ATL/BTL por seniority, esqueletos de email adaptables y cadencias de secuencia. Usar cuando el usuario pregunte 'cuánto personalizo', 'personalización a escala', 'primera línea del email', 'hooks', 'qué research hago del prospect', 'plantillas de cold email', 'cadencia', 'cada cuánto mando follow-ups', 'secuencia de toques', o al preparar los mensajes de una campaña outbound. Complementa el paso de personalización batch de Yalc (rol + empresa + señal verificada) y define el tier profundo para el futuro Investigador. NO usar para elegir el trigger (ver b2b-sales-triggers) ni definir la audiencia (ver b2b-targeting-playbook).

Catálogo accionable de 137 sales triggers para outbound B2B, destilado de ColdIQ (metodología Flip the Script). Cada trigger responde: qué señal lo dispara, cómo detectarla y qué ángulo de mensaje activa. Usar cuando el usuario pregunte 'con qué excusa/razón contacto a este lead', 'qué trigger uso', 'por qué escribirle ahora', 'sales triggers', 'buying signals', 'señales de compra', 'campaña signal-based', 'win-back', 're-engagement de closed-lost', 'a quién priorizo del inbound', o esté montando outbound y necesite elegir la premisa de contacto. NO usar para definir el ICP (ver b2b-targeting-playbook) ni para decidir el nivel de personalización o redactar la secuencia (ver b2b-personalization-depth).

Playbook de targeting e ICP para outbound B2B, destilado de las master skills de list building y ABM de ColdIQ. Define y refina audiencias: ICP en 3 capas, scoring 0-100 con tiers A-D, criterios de exclusión, lookalikes desde mejores clientes, dimensionado de lista (revenue reverse-engineering) y mapeo del comité de compra. Usar cuando el usuario pida 'definir el ICP', 'ideal customer profile', 'a quién apuntamos', 'targeting', 'audiencia de outbound', 'segmentar la lista', 'scoring de cuentas', 'priorizar cuentas', 'tiers ABM', 'target account list', 'criterios de exclusión', 'cuántas cuentas necesito', o antes de lanzar cualquier campaña outbound nueva. NO usar para elegir el trigger de contacto (ver b2b-sales-triggers) ni para redactar mensajes o decidir personalización (ver b2b-personalization-depth).

Planifica por chat una búsqueda de creators para Partnerships (SAN-79). Conversación en el chat global de Sancho: recomienda sectores/redes/tiers según el contexto del cliente, itera con el usuario y produce un PLAN estructurado (plan-card). Al confirmar, crea la búsqueda: campaign type=Partnerships en Yalc + tarea Outreach madre + runner de discovery encolado. Usar cuando: 'crear nueva búsqueda', 'busca creators', 'lanza un discovery de influencers', 'programa de creators', o desde el botón 'Crear nueva búsqueda' del tab Encuentra (Outreach). NO usar para: outreach B2B clásico (outreach-playbook), construir secuencias (outreach-sequence-builder), ni ejecutar el scraping (discovery-search-runner).

Design a complete acquisition metrics plan for any business type. Classifies the business into archetypes (SaaS/App, Fintech, Marketplace, E-commerce/D2C, Lead-to-Sale, Hybrid), defines activation events, builds 4-level metrics hierarchy, sets benchmarks, creates review cadence, and generates a personalized Excel tracking template. Reads company-context, budget, ecps from Context Lake. Writes metrics-plan.md to brand/. Use when onboarding a new client, launching a product, restructuring metrics tracking, or user says design metrics plan, what should we measure, acquisition KPIs, metrics setup, tracking plan. Do NOT use for diagnosing existing metrics (use diagnose) or for designing experiments (use design-experiment).

Connect a client to an external API (GA4, GSC, Meta Ads, HubSpot, Stripe, etc.) via Mission Control. Use when user says connect, conecta, conectar, link, vincular, integrar followed by an API name or service. Also triggers on "quiero conectar", "necesito conectar", "configura la API de", "añade", "enlaza". NEVER ask for credentials in chat — always redirect to the Mission Control connect page.

Ejecuta el plan de una búsqueda de creators (Partnerships · SAN-79): scrapea candidatos con las tools MCP de ScrapeCreators según el plan (sectores/redes/tiers), comprueba el repeat de competidores vía ad-library, normaliza candidatos (handle, red, followers, ER, señales) y los ingesta en Yalc, donde entran como Leads scoreados (quality score real con calc-creator-core) según el qualification_mode de la campaign. Usar cuando: hay búsquedas con runner 'queued' (job/cron de discovery), 'ejecuta la búsqueda X', 'corre el runner de discovery'. Modo fixture disponible (sin ScrapeCreators) para tests y demos. NO usar para: planificar la búsqueda (discovery-plan-builder) ni outreach B2B (apollo/company-finder).

Build cold outreach sequences that get replies. Use when the user wants to write cold outreach emails, prospecting emails, cold email campaigns, sales development emails, SDR emails, or multi-channel outbound sequences. Also use when the user mentions 'cold outreach,' 'prospecting email,' 'outbound email,' 'email to leads,' 'reach out to prospects,' 'sales email,' 'follow-up email sequence,' 'nobody is replying to my emails,' 'how do I write a cold email,' 'cold email,' 'LinkedIn outreach,' 'outbound sequence,' or 'book more meetings.' Covers subject lines, opening lines, body copy, CTAs, personalization, multi-touch follow-up sequences, and multi-channel (email + LinkedIn + video). For warm/lifecycle email sequences, see email-sequence. For sales collateral beyond emails, see sales-enablement.

Use BEFORE creating any branch, commit, PR, or release in sanchocmo-openclaw — the repo's mandatory git conventions. Triggers whenever you are about to `git checkout -b`, name a branch, write a commit message, open a PR (`gh pr create`), pick a merge method, or cut/ship a release. Covers branch naming (`<author>/san-<n>-<desc>`), Conventional Commits, squash-to-`main`, the single-trunk model, and the release + hotfix flow.

Context pack for fast, read-only review of private Growth4U HTML documents.

Operate YALC/GTM-OS from Sancho through deterministic outbound workflows: provider checks, campaign preparation, verified personalization, immutable batch approval, LinkedIn/Unipile execution, email/Instantly handoff, status, and reporting. Use when the user asks Sancho to prepare, review, send, inspect, or troubleshoot outbound.

Operate Alarife Payload from Sancho. Use when creating, importing, exporting, editing, previewing, publishing, or auditing sites/pages in Alarife Payload; when user says "exporta a Alarife", "Alarife Payload", "Payload", "cambia el diseño", "importa un sitio", "crea una landing", "publica", or "preview".

Run Apify Actors (web scrapers, crawlers, automation tools) and retrieve their results using the Apify REST API with curl. Use when the user wants to scrape a website, extract data from the web, run an Apify Actor, crawl pages, or get results from Apify datasets.

Interact with Apollo.io REST API (people/org enrichment, search, lists).

Brand coherence verification de cualquier output de marca (post, blog, ad, email, slide, página, thread) contra la Foundation. Aplica 4 dominios: voice & tone (siempre), messaging consistency (siempre que haya claims), visual identity (cuando aplique), SEO basics (cuando aplique). Devuelve veredicto PASS / NEEDS REVISION / MAJOR ISSUES con score y action list priorizada. Skill de Sansón — invocada solo vía dispatch (`sessions_send` con `**Tipo**: brand-check`), no user-invocable. Use when: pipeline ha generado un draft y antes de publicar, cualquier sancho-manager request con tipo brand-check, o tras editar un asset existente. NOT for: fact-checking de datos numéricos (eso es qa-bot vía deep-research Phase 6), copy-editing prosaico (eso es copywriting), generación de contenido (skills de canal como social-writer/seo-content/newsletter).

Define brand voice, tone, vocabulary, Do/Don't rules, and AI Brand Kit. Use when: codifying how the brand speaks for consistent content creation. Two modes: Quick (Layer 0, ~30min, from URL or 5 questions) produces Voice Snapshot; Full (Layer 4, ~2-3h, after ECPs + positioning) produces complete Voice Guide + AI Brand Kit + Per-ECP and Per-Channel adaptations. Quick mode MANDATORY before any content. Full mode adds ECP tone shifts, channel guidance, vocabulary rules, Do/Don't library, and 5-check Voice Test. NOT for: visual identity (use visual-identity), content creation (use content skills), or positioning strategy (use positioning-messaging).

Pick best marketing channels to activate.

Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. Also handles migrating existing Claude API code between Claude model versions (4.5 → 4.6, 4.6 → 4.7, retired-model replacements). TRIGGER when: code imports `anthropic`/`@anthropic-ai/sdk`; user asks for the Claude API, Anthropic SDK, or Managed Agents; user adds/modifies/tunes a Claude feature (caching, thinking, compaction, tool use, batch, files, citations, memory) or model (Opus/Sonnet/Haiku) in a file; questions about prompt caching / cache hit rate in an Anthropic SDK project. SKIP: file imports `openai`/other-provider SDK, filename like `*-openai.py`/`*-generic.py`, provider-neutral code, general programming/ML.

Comic book UI: halftone, panels, bubbles.

Find ICP-matching companies.

3-Lens competitor analysis: Autopercepción, Terceros, Consumidores. Use when: researching competitors, building battle cards, mapping competitive landscape. Discovers competitors, profiles each (scraping + deep research), runs 3-lens analysis, produces battle cards and landscape map. NOT for: self-analysis (use self-intelligence), market sizing (use market-intelligence).

Get verified emails, phones, socials.

Repurpose content across platforms.

Plan editorial calendar and cadence.

Populates Content Engine config files with client-specific data and writes a narrative setup.md explaining what was set up and which crons consume it. The infrastructure (folders, YAML templates, cron jobs) already exists — this skill only FILLS IN the client-editable fields and DOCUMENTS what was done.

Generate or attach media to a Content Engine draft. The ONLY legitimate way for an agent to put images, carousels, or PDFs into a draft's media[] so the Mission Control MediaPanel renders them. Use whenever a writer skill (newsletter, social-writer, seo-content, instagram-content, content-atomizer) needs a hero image, header, carousel, or any visual asset for the approved draft. NEVER edit frontmatter.media by hand.

Define and maintain Content Pillars for a client. Uses a 5-layer methodology: Content Tilt → Pain-Based+JTBD → BOFU-First Prioritization → Topic Cluster Structure → Playground Tagging. Outputs readable markdown with rich context per pillar. Pillars = TOPICS the brand will own. POV is NOT decided here — POV lives at piece level during angle/clarify.

Generate and update the Content Playbook for a client. Defines content pillars, hooks, formats, and cadence based on Atalaya intelligence, Trust Engine SEO data, and Foundation context.

When the user wants to plan a content strategy, decide what content to create, or figure out what topics to cover. Also use when the user mentions "content strategy," "what should I write about," "content ideas," "blog strategy," "topic clusters," "content planning," "editorial calendar," "content marketing," "content roadmap," "what content should I create," "blog topics," "content pillars," or "I don't know what to write." Use this whenever someone needs help deciding what content to produce, not just writing it. For writing individual pieces, see copywriting. For SEO-specific audits, see seo-audit. For social media content specifically, see social-content.

Daily insights from comms and meetings.

Find decision makers at target companies.

Multi-source research in 7 mandatory phases (SCOPE → SOURCES → EXTRACT → FRAMEWORK → DETAIL → QA → DELIVER) with structured output, entity-level breakdowns, and mandatory qa-bot verification. Use when user says 'deep research', 'investiga [topic]', 'research [topic] for [client]', 'análisis de mercado', 'competitive analysis', 'benchmark [topic]', 'profundizar', 'investigar más', or needs a comprehensive market/product/regulatory investigation. Each entity gets the same template (no asymmetric coverage). Minimum 10 unique sources, 3-5 web_search per section in Spanish AND English, every claim sourced. Optional Phase 2b social pulse via last30days for community/recency dimensions. NOT for: quick factual lookups (use WebSearch directly), or content generation (use seo-content / social-writer / newsletter).

Crear o actualizar el DESIGN.md (source-of-truth del visual identity del brand). Skill propia de Maese Pedro, forkeada de `design-brief` (Open Design) + discovery interactivo heredado de `visual-identity` legacy. Genera DESIGN.md siguiendo schema OD de 9 secciones, enriquecido con patrones Sancho (logo color rules, social specs por canal, color use semántico, illustration discipline, anti-pegote, tamaños mínimos de texto). Triggers: "design system", "visual identity", "DESIGN.md", "brand visuals", "look and feel", "design tokens", "actualizar paleta", "brand discovery".

Write direct-response copy that converts: ads, landing pages, emails, sales pages. Pipeline: Context → Format Selection → Draft with DR techniques → Humanize → Review. Applies 13 proven techniques (headlines, curiosity gaps, slippery slide, pain quantification, founder stories, testimonials, CTAs). Anti-AI-detection built in.

OPTIONAL pillar — test ECP assumptions before heavy execution. Use when: 4+ weeks timeline, multiple ECPs to choose between, high execution cost, pre-launch. Maja Voje frameworks: Assumption Mapping, Method Selection, MVI. Methods: interviews, landing page smoke tests, waitlists, presale, MVI. Skip if: timeline < 4 weeks, single obvious ECP, low execution cost, client prefers validate-through-execution.

Analyze existing customer/CRM data: RFM segmentation, behavioral clustering, churn patterns, LTV analysis, upgrade patterns. OPTIONAL pillar — skip if pre-launch, no CRM access, or <50 customers. Use when: client has >50 customers with CRM data available and wants data-driven ICP refinement, churn reduction, or segment optimization. Produces Champions profile, RFM segments, churn patterns, LTV by segment, and actionable recommendations for ICP/positioning/content. NOT for: pre-launch companies, assumption-based ICP (use niche-discovery-100x), or general market research (use market-intelligence).

Orquesta la Foundation v3.0: 7 secciones, 8 layers, gate checks con requires/enriches_with. Flujo: Kickoff (1 skill, company-brief.current.md) → Trust Score (opcional, auto-arranca tras Company Brief) → Full Foundation (9 skills individuales) → Metrics Setup → Strategic Plan. Al aprobar un pilar, ejecuta automáticamente el siguiente. Leer pillar-registry.md para detalle de cada pilar.

Define y gestiona las VOCES de Founder-Led Content. Una voz = una persona en una red (p.ej. Alfonso·LinkedIn, Martín·LinkedIn, Alfonso·X). Captura quién + en qué red + handle + cadencia + cuenta de publicación (metricool_profile_id) + tono, y las escribe como profiles[] bajo el canal correcto en cadence-config.yml. Úsala SIEMPRE que haya que crear, añadir o editar voces founder-led: 'añadir voz', 'añade a Martín en X', 'da de alta una voz/founder en LinkedIn o Twitter', 'configura quién publica founder-led', o conectar la cuenta de publicación de una voz. Se invoca standalone o desde content-engine-setup (Step 4, cadencia).

Create stunning, animation-rich HTML presentations from scratch or by converting PowerPoint files. Use when the user wants to build a presentation, convert a PPT/PPTX to web, or create slides for a talk/pitch. Helps non-designers discover their aesthetic through visual exploration rather than abstract choices.

Design qualification and conversion funnels.