Synthèse automatique
Un brief de contexte persistant qui décrit Ed Dale, sa mission, son modèle de livraison en triangle et son vocabulaire propriétaire pour que l'agent travaille sur ses projets avec le bon cadre.
Points forts
- Condense en un seul fichier l'identité du client, son entreprise, son parcours et sa promesse centrale.
- Fournit un filtre de décision explicite (la « Prime Directive » : premier gain IA en 30 jours, puis effet cumulé) utilisable pour arbitrer chaque idée.
- Cadre le ton grâce aux modèles mentaux nommés (Halbert, King, Zinsser) et à un lexique maison réutilisable.
- Délimite clairement ce qu'il ne faut PAS construire (SaaS, plateformes techniques complexes), ce qui évite les dérives de périmètre.
Limites
- Entièrement spécifique à une seule personne : sans valeur en dehors des projets d'Ed.
- Le fichier est tronqué : le tableau de terminologie s'arrête au milieu de l'entrée « Prompt Wrangling™ ».
- Aucun échantillon de rédaction réel n'est fourni, donc imiter sa voix reste approximatif malgré les références citées.
À charger dès qu'une tâche touche à la rédaction, au positionnement ou à la priorisation de projets pour Ed ou ses programmes de coaching.
À éviter pour du travail technique générique ou pour tout client extérieur à l'écosystème d'Ed, où ce contexte nuirait plus qu'il n'aiderait.
Analyse de sécurité
SûrThe skill is purely informational and declarative, providing context about a person and their business. It declares only read and write tools (Read, Glob, Grep, WebSearch, WebFetch, Write, Edit) and contains no instructions to execute commands, modify system files, exfiltrate data, or perform any destructive or unsafe actions.
Aucun point d'attention détecté
Comment cette analyse est faiteExemples
Load the mission-context skill and write a launch email for the AI Amplified Coaching Powerhouse program. Use Ed's voice, open with a Halbert-style hook, and anchor it to the promise of a first AI win in 30 days.Here's an idea for a client dashboard with user accounts and billing. Run it through the Prime Directive from Ed's mission context — does it get coaches their first win in 30 days and compound? Tell me whether to build it or kill it.Draft this week's Little Blue Report issue explaining the Capabilities X-Ray, using Ed's terminology and the RED/GREEN/BLUE triangle framing.name: mission-context description: Provides context about Ed, his mission, terminology, and business. Use when working on Ed's projects, coaching business, AI Powerhouse content, BlackBelt work, or The Little Blue Report. Also use when Ed asks you to understand his business or speak in his voice. allowed-tools: Read, Glob, Grep, WebSearch, WebFetch, Write, Edit
Mission Context - Ed's Operating Context
This skill gives you everything you need to work effectively with Ed. Load this context when working on his projects.
WHO ED IS
Ed Dale - Vibe Coder, world-class copywriter, and coach. Founder of the 30 Day Challenge, The first ever online challenge. Author of the Book Your First Dollar
Company: Neon Marillion Pty Ltd
Background:
- Senior coach at BlackBelt coaching program
- Created the AI Amplified Coaching Powerhouse program
- Publishes "The Little Blue Report" on Substack
- Thinks in headlines, hooks, and frameworks
Superpower: He knows what good looks like. He doesn't need to know how to build it - he needs AI to translate his vision into working reality.
Mental models come from: Gary Halbert, Bond Halbert, Stephen King, William Zinsser. Use copywriting and coaching analogies when explaining things.
THE MISSION
The One-Sentence Strategy (Prime Directive)
"First AI win in 30 days. Then compound."
This is the filter for every decision. If it doesn't help coaches get their first win fast or build compounding leverage, it's a distraction.
Agent check: Before building anything, ask:
- Does this help them win in 30 days?
- Does this compound over time?
- If no to both → reconsider.
The Full Mission
Transform established coaches ($30k+/month) into AI Amplified Coaching Powerhouses - safely, sanely, and in a way that feels like them.
The core promise:
Within the first 30 days, you will have your first AI leverage tool deployed live in your business, with a clear roadmap for what to build next.
What Ed is NOT building:
- SaaS companies
- Complex tech platforms
- Anything that requires becoming a tech founder
What Ed IS building:
- Leverage assets that do the lifting
- Skills that can be reused and shared
- Systems that compound over time
THE TRIANGLE MODEL
Ed's delivery framework has three layers:
RED - The Problems (What clients feel)
- Everyone seems to be moving faster with AI
- Every tech project turns into a nightmare sinkhole
- Lots of AI toys, AI slop, and no real lift
GREEN - The Milestones (What they achieve)
- Leverage opportunities mapped - Know where the wins are
- Leverage assets live - Actually deployed and working
- Leverage engine amplification - System keeps improving itself
BLUE - The Projects (How we get there)
Milestone 1: Leverage Opportunities Mapped
- Capabilities X-Ray (scan the business for hidden opportunities)
- Plays Shortlist (prioritized by riskiest assumption first)
- Upgrade Roadmap (90-day sequence with clear success signals)
Milestone 2: Leverage Assets Live
- Prompt Whispering™ Prototypes (risk-free proving grounds)
- Deployed Leverage Assets (in live workflows, team-safe)
- Leverage Skunkworks (controlled factory with kill rules)
Milestone 3: Leverage Engine Amplification
- Founder AI Lab (test, break, validate what matters)
- Learn & Adapt Cycles (usage → insight → upgrades)
- Compound & Retire Protocols (scale what works, kill what doesn't)
ED'S TERMINOLOGY (Use These)
| Term | Meaning | |------|---------| | Capabilities X-Ray | Scanning a business for leverage opportunities | | Plays Shortlist | Prioritized task list, riskiest assumption first | | Leverage Asset | Any tool that does the lifting | | Kill rules | When to stop and pivot vs. keep pushing | | Ship This = Win Day | The only metric that matters | | Prompt Whispering™ | Building AI tools through conversation | | Prompt Wrangling™ | Debugging and refinement | | Upgrade Roadmap | 90-day sequence of experiments | | Leverage Skunkworks | Controlled factory for building/testing assets | | Founder AI Lab | Always-on testing of what's working in AI |
HOW TO TALK TO ED
Translation Layer (Tech → Copywriter Speak)
| Technical | What Ed Hears | |-----------|---------------| | Component | Reusable template (like a swipe file element) | | Props | Variables you pass in (like merge tags) | | State | What the page remembers (like cookies) | | API | Backend conversation (like a database query) | | Deploy | Publish live (like hitting "send") | | Refactor | Editing for clarity without changing meaning | | Debug | Finding why the copy isn't converting | | Dependencies | Tools in your toolbox | | Repository | Master swipe file for this project |
Communication Style
- Short, punchy responses
- Use analogies from copywriting and coaching
- Explain the "why" in plain English
- No unnecessary jargon
- Celebrate wins - shipping matters
THE IRON RULES
- DO NOT BE LAZY. Find root causes. No temporary fixes.
- MAKE FIXES SIMPLE. Minimum code impact. Zero new bugs goal.
- ALWAYS EXPLAIN THE WHY. Plain English. Use analogies.
TECH STACK
The Vibe Coder Rule
"If I want it to think smart, use Python. If I want it to look cool, use React."
Hardware (Apple House)
- Mac Studio M4 Pro: Primary workstation
- iPhone 16 Pro Max: Mobile
- Apple Vision Pro: First generation
- iPad Pro 13-inch (M4): Tablet
Development & Deployment
- Claude Code: Primary AI coding partner (CLI)
- Antigravity IDE: Google's IDE with Claude extension
- Claude.ai: Web interface for conversations & skills
- GitHub: Version control, skill distribution
- Vercel: Hosting & deployment (React/Next.js)
- Google Cloud Run: Deployment (Google stack)
- Google AI Studio: AI development & testing
Content & Knowledge
- Obsidian: Daily notes, content pipeline, knowledge base
- Substack: The Little Blue Report publication
- MacWhisper: Dictation and transcriptions
- Descript: Video editing
- DaVinci Resolve: Video editing (latest)
Daily Drivers
- Arc: Primary browser
- Comet: Secondary browser
- Apple Mail: Email client (Claude Code can draft/send via AppleScript)
- Google Calendar: Calendar
Languages
- Python: When it needs to think (AI, automation)
- React/Next.js: When it needs to look good (UI)
- Markdown: Everything else (skills, docs, notes)
AI Tools
- Claude: Primary AI partner (Code + .ai)
- ChatGPT: Conversations, brainstorming
CRM & Marketing Automation
- GoHighLevel (GHL): CRM, workflows, lead capture, landing pages
- Ed uses CMM (Coach Marketing Machine) which is a white-label of GHL (identical functionality)
- When writing content, refer to "GHL" for audience clarity (most won't know CMM)
- Integration strategy: Direct API integration - Claude builds lead magnet sites that connect directly to GHL API for CRM updates and workflow triggers (no Zapier/n8n middleware)
Socials & Platforms
| Platform | Handle | Usage | |----------|--------|-------| | YouTube | @eddale | Weekly long-form value content (key focus going forward) | | Instagram | @eddale | Daily posting: 4 days value/engagement, 1 day lead magnet handraiser (via GHL) | | Twitter/X | @Ed_Dale | Read daily, rarely post | | Facebook | Business + Personal pages | Experimenting with Instagram carousel reposts | | LinkedIn | Neon Marillion Pty Ltd | Recently refreshed | | Linktree | linktr.ee/eddale | Link hub | | Substack | The Little Blue Report | Primary newsletter |
Agent Capabilities (What Claude Code Can Do)
- Email (Send): Draft and open in Mac Mail via AppleScript (review before send) ✓
- Email (Search): Limited - AppleScript too slow for large mailboxes, Spotlight broken since Catalina. Future: Python IMAP or Gmail API
- Files: Read/write to Obsidian vault and GitHub repos
- Web: Fetch and analyze URLs, search the web
- Git: Full version control, commits, pushes
- Shell: Run scripts, package skills, system automation
SKILL ARCHITECTURE
Ed's skills follow a three-layer pattern:
The Three Layers
| Layer | What It Is | Who Maintains | |-------|------------|---------------| | Tools | Python scripts, APIs, Bash | Hidden inside skills | | Skills | SKILL.md with instructions | Edit markdown to change | | Agents | Orchestrators pulling skills | Reference skills, not tools |
Skills Own Their Tools
- Python scripts live in
skills/<name>/tools/ - SKILL.md tells Claude how to run them
- Agents reference skills, not tools directly
Context Propagation
When writing content:
- Audience should be established early (ask if not specified)
- Pass context to downstream skills (hook-stack, slop-detector)
- Store in article frontmatter for visibility
The Newsletter Pipeline
- Input (YouTube URL, topic, experience)
- Establish audience (ask if not specified)
- Extract/analyze content
- Generate headlines with audience context
- Score with Hook Stack (12+ required)
- Draft article in Ed's voice
- Run through slop detector
- Save to Zettelkasten with metadata
Runtime Differences
Skills must account for which Claude runtime executes them:
| Capability | Claude Code (CLI) | Claude.ai (Mac/Web) | |------------|-------------------|---------------------| | Filesystem | Full access | Sandboxed | | External APIs | Yes (curl, WebFetch) | Blocked by sandbox | | Bash/Python | Yes | No | | MCP Servers | Supported | Primary method for external data | | Best for | Automation, building | Conversation, local-tool skills |
Design implications:
- Claude Code skills can use deployed APIs (Vercel, etc.)
- Claude.ai skills should either:
- Include all tools locally (
tools/folder with scripts) - Use MCP servers for external data access
- Accept manual paste as fallback
- Include all tools locally (
Example: youtube-processor has both local Python script (Claude Code) and Vercel API (doesn't work in Claude.ai due to sandbox - future MCP solution planned).
CURRENT FOCUS AREAS
Powerhouse Lab (This Repo)
Ed's Leverage Skunkworks where AI tools get built, tested, and validated.
Lifecycle:
prototypes/ → skills/ → [Own Repo]
↓
archive/ (kill rule applied)
The Little Blue Report
Ed's Substack newsletter. Educational content drawn from daily experiences.
BlackBelt Coaching
Ed's a coach in Taki Moore's high-end coaching program.
AI Amplified Coaching Powerhouse
Ed's flagship coaching program for implementing AI into successful coaching businesses.
WHEN TO USE THIS SKILL
Activate this context when:
- Working on any Ed project
- Writing content in Ed's voice
- Building tools for the coaching business
- Discussing strategy or positioning
- Creating materials for clients
- Anytime you need to understand Ed's world
UPDATING THIS CONTEXT
Ed can update this skill anytime by saying:
- "Update my mission context to include..."
- "Add this to my terminology..."
- "Change my current focus to..."
When Ed requests changes, edit this SKILL.md file directly.
Du fichier au premier essai
Le téléchargement contient exactement le contenu affiché sur cette fiche. Cette copie peut différer de la dernière version du dépôt source.
Consulter la source et ses fichiers annexesCe téléchargement contient uniquement SKILL.md. Les scripts, références, modèles et dépendances éventuels ne sont pas inclus. Récupérez les fichiers nécessaires depuis la source en conservant leur arborescence avant de poursuivre.
1. Télécharger le fichier
Télécharger SKILL.mdDepuis la racine d’un projet de test, enregistrez le fichier sous .skill-downloads/mission-context-ed-dale-msfn18xj/SKILL.md, ou utilisez la commande suivante. Ce dossier sert à examiner la skill avant son installation.
Commande pour macOS/Linux ou WSL : un shell POSIX et curl sont nécessaires. Sous Windows sans WSL, utilisez le bouton de téléchargement et créez le dossier manuellement.
mkdir -p '.skill-downloads/mission-context-ed-dale-msfn18xj' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/mission-context-ed-dale-msfn18xj/download' --output '.skill-downloads/mission-context-ed-dale-msfn18xj/SKILL.md.part' && mv '.skill-downloads/mission-context-ed-dale-msfn18xj/SKILL.md.part' '.skill-downloads/mission-context-ed-dale-msfn18xj/SKILL.md'2. Vérifier les prérequis
Lisez SKILL.md avant de l’utiliser. Si le fichier cite des scripts ou services externes, examinez-les et préparez les accès nécessaires dans votre environnement de test.
Outils déclarés par l’auteur (à vérifier dans votre agent) : Read, Glob, Grep, WebSearch, WebFetch, Write, Edit
Aucune section de prérequis reconnue dans ce fichier. Cela ne signifie pas qu’il est autonome : les bibliothèques, outils et accès peuvent être cités ailleurs dans le contenu ou dans le dépôt source.
Les plateformes de la fiche sont des indications du catalogue. Les chemins ci-dessous suivent la documentation des outils ; ils ne prouvent pas que cette skill y a été exécutée. Aucune version minimale de ces outils n’est attestée pour cette fiche.
3. Charger la skill
Après vérification, copiez le dossier complet dans le projet avec l’une des commandes ci-dessous, puis ouvrez une nouvelle session de l’outil. La commande s’arrête si un dossier du même nom existe déjà : comparez les fichiers avant de le remplacer.
Claude Code
test -f '.skill-downloads/mission-context-ed-dale-msfn18xj/SKILL.md' && test -s '.skill-downloads/mission-context-ed-dale-msfn18xj/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/mission-context' && cp -R '.skill-downloads/mission-context-ed-dale-msfn18xj' '.claude/skills/mission-context'OpenSkills
OpenSkills doit déjà être installé. La procédure locale install/read est contrôlée avec OpenSkills 1.5.0, qui exige Node.js ≥ 20.6.0. Aucun slug du catalogue n’est résolu automatiquement.
Installation officielle d’OpenSkillsopenskills install './.skill-downloads/mission-context-ed-dale-msfn18xj'openskills read 'mission-context-ed-dale-msfn18xj'La lecture doit afficher le contenu du fichier installé et son dossier de base. Elle ne lance ni la skill ni ses dépendances.
Pour un agent qui lit AGENTS.md, synchronisez ensuite l’index. Cette commande modifie AGENTS.md : relisez le diff avant de l’utiliser.
openskills syncMode manuel ou autre outil : joignez SKILL.md à une conversation, ou ouvrez-le dans un agent capable de lire les fichiers. Ajoutez les ressources annexes nécessaires. Il n’existe pas de dossier universel reconnu par tous les outils.
4. Faire un premier essai
Utilisez une entrée courte et fictive : un endpoint d’API, un petit document ou un brief d’interface. Dans une nouvelle conversation, joignez le fichier ou donnez son chemin à l’agent, puis adaptez cette demande :
Lis .skill-downloads/mission-context-ed-dale-msfn18xj/SKILL.md et applique ses instructions à cet exemple : [décrire une entrée de test]. Indique les fichiers, dépendances ou accès qui manquent. Produis le résultat demandé par la skill sans inventer de données ni exécuter de script non examiné.Résultat attendu : une sortie correspondant aux instructions de la skill, ou une liste précise des prérequis manquants. Une simple confirmation de lecture ne valide pas le premier usage. Si une ressource externe est indisponible, ce parcours reste à tester.
Fragments d'écriture
Contenu
Extrayez des fragments d'écriture bruts via un entretien intensif, sans imposer de structure. Capturez idées, citations et mots porteurs pour une utilisation ultérieure.
Rédaction UI et audit
Contenu
Rédigez ou auditez les textes d'interface (boutons, erreurs, états vides, infobulles) dans le monorepo, en suivant la source de vérité du design system.
Extraction MIDI/score
Contenu
Extrait le MIDI ou la partition d'une production musicale existante via l'outil MCP CogMusic et confirme la disponibilité du fichier.