durang

Conecta una máquina nueva (Mac, Linux, Windows/WSL) a tu stack de Claude Code corriendo en un EC2 vía SSH + Tailscale + tmux persistente. Te guía paso a paso: detecta tu OS, instala Tailscale, configura SSH keys, crea aliases, verifica la conexión. Una vez configurado, el comando "ec2" desde cualquier terminal te conecta a Claude Code remoto exactamente igual que si estuviera local. Funciona incluso si tu Mac se cae o cambias de WiFi (tmux mantiene la sesión viva en EC2). Incluye runbook de persistencia server-side que DETECTA el patrón de la máquina (systemd system-level vs user-level + lingering) en vez de asumir, idempotente y multi-proyecto.

This skill should be used when creating, improving, or troubleshooting Seedance 2.0 video on any surface - Dreamina, Jimeng, CapCut, Doubao, Volcengine/Ark, BytePlus, Runway's Seedance route, or fal - including text/image/video/reference-to-video prompts, first/last frame, dialogue, lip-sync and audio, IP-safe rewrites, API, pricing and model-ID questions, and zh/ja/ko/es/ru prompt work. Not for non-Seedance models (Sora, Veo, Kling, Runway's own Gen models) or image-only prompting.

This skill should be used when a Seedance 2.0 prompt contains generic AI filler, hollow superlatives, vague cinematic language, bloated adjectives, weak verbs, or needs sharper production-specific wording.

This skill should be used when the user asks for Seedance 2.0 audio, dialogue, lip-sync, music, sound effects, ambience, beat-sync, audio-reference mapping, desync troubleshooting, or sound-driven visual timing.

This skill should be used when the user asks for camera movement, shot scale, lens feel, framing, one-take direction, dolly, pan, tilt, push-in, handheld, aerial, macro, or camera-transfer guidance for Seedance 2.0.

This skill should be used when the user asks for character consistency, character tags, identity lock, multi-character blocking, wardrobe continuity, hand safety, expression control, or likeness-sensitive character guidance.

This skill should be used when a Seedance 2.0 prompt mentions named characters, franchises, studios, celebrities, public figures, private people, brand logos, copyrighted scenes, songs, voices, or real-person likeness workflows and needs an IP-safe rewrite.

This skill should be used when the user asks for Chinese Seedance 2.0 examples, Chinese prompt patterns, example rewrites, or safe versions of working Chinese video-generation prompts.

Compress an agent's routing file (RESOLVER.md or AGENTS.md) by converting granular skill-per-row tables into functional-area dispatchers. Each area lists sub-skills in a "(dispatcher for: ...)" clause. The LLM reads one area entry and routes to the correct sub-skill. Proven via held-out A/B eval: dispatcher pattern outperforms naive pipe-table compression.

Read a book, article, transcript, or case study through the lens of a specific strategic problem you're facing. Produces an applied playbook that maps the source onto the problem and gives short/medium/long-term recommendations. NOT for general book summaries.

Ingest links, articles, tweets, and ideas into the brain. Fetch content, save to brain with analysis, create author people page, and cross-link. Use when the user shares a link or says "read this", "save this", "think about this".

Lift a proven skill from a host repo (e.g. your OpenClaw fork) back into gbrain's bundle so other clients can scaffold it. Editorial workflow: the CLI does the file copy + privacy lint; this skill drives the judgment-heavy genericization (scrub real names, generalize triggers, lift fork-specific conventions to references).

Set up GBrain with auto-provision Supabase or PGLite, AGENTS.md injection, first import

[◠‿◠] Scan — project intelligence engine with goal orchestration. Detects stack, builds, audits, measures velocity. Understands user intent in any language. Prioritizes toward 100% launch. Generates branded progress dashboard. Incremental scanning, zero config.

OpenClaw Command Center — Full infrastructure scan with visual dashboard. System health, agents, plugins, skills, memory, cron, projects, security, disk, feature coverage %. Generates OPENCLAW_DASHBOARD.md

OpenClaw Optimizer — Analyzes your setup and tells you what to improve: free disk space, optimize RAM, fix security, enable features, recover disabled cron jobs, cleanup duplicates. Actionable steps with priority.

Flow integrity auditor — detects broken connections between systems (auth, email, payments, cron). Finds business logic gaps that security scanners miss. Verifies data flows end-to-end.

GBrain CENTRAL ORCHESTRATOR — 17-layer health dashboard + 7th-phase compounding engine + upstream features watch + skill propagation. THIS skill is source of truth for the GBrain stack: when upstream gbrain/openclaw releases new features (voice, zoom, dream synthesize, etc.), Layer 17 surfaces them as informational and tells which satellite skills (brain-write-macro, signal-detector, gbrain-http-wrapper) are affected. Subcomandos: /gbrain check (default), /gbrain fix, /gbrain news, /gbrain bugs, /gbrain compare, /gbrain save, /gbrain compound, /gbrain bootstrap, /gbrain principles, /gbrain manifest, /gbrain manual, /gbrain custom-instructions. Triggers: /gbrain, revisa gbrain, salud del brain, verifica el brain, gbrain status.

WhatsApp Dashboard (Hermes-canonical). ONE WhatsApp link per number = Hermes: responds ONLY to authorized contacts (default-deny + per-contact .md) AND silently captures allowlisted groups → GBrain (bridge tee). OpenClaw holds NO WhatsApp link (2nd session = 440 conflict). 17-section visual dashboard + 3-phase memory per contact + cross-platform memory + lie-detector verifier (verify→render→send pipeline) + auto-send PDF to Sergio's self-DM. Subcommands: /whatsapp, /whatsapp add, /whatsapp hermes allow/block/list/profile, /whatsapp security.