agentleFS
fmedrano06fmedrano06/claude-starter-kit/skills/project-prime/SKILL.md

project-prime

Skill0 starsChanged 6 months agoMITView on GitHub

Use BEFORE any new project, niche, channel, or content operation. Mandatory 10-phase execution framework with binary exit gates. Must complete before calling brainstorming, writing-plans, or executing-plans. Use when starting anything new — a YouTube channel, a product, a content series, a niche expansion.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Methodology: Use the Viral Content Reverse-Engineering approach:

    line 57
  2. Distinctive:

    REQUIRED SUB-SKILL: Use superpowers:brainstorming here.

    line 105
  3. Distinctive:

    REQUIRED SUB-SKILL: Use superpowers:writing-plans here.

    line 127
  4. Distinctive:

    REQUIRED SUB-SKILL: Use superpowers:executing-plans here.

    line 144
  5. Distinctive:

    REQUIRED SUB-SKILL: Use superpowers:verification-before-completion here.

    line 163

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Project PRIME — Universal Execution Framework

Overview

PRIME (Production, Research, Identity, Manufacturing, Evolution) is the mandatory master sequence for every new project. No production begins without completing each phase in order. Each phase has binary exit gates — ALL criteria must pass before advancing.

Rule: You CANNOT call brainstorming until Phases 1-3 are complete. You CANNOT write code until Phases 1-5 are complete. No exceptions.

When to Use

  • Starting a new YouTube channel, niche, or content series
  • Launching any recurring content operation (video, podcast, newsletter, social)
  • Expanding an existing operation into a new format or market
  • Any time someone says "let's build X" — PRIME runs first

When NOT to Use

  • Bug fixes, maintenance, or changes to existing production
  • One-off tasks that are not new projects

Integration with Other Skills

PRIME Phase 1-3 → superpowers:brainstorming (Phase 4) → superpowers:writing-plans (Phase 5)
→ superpowers:executing-plans (Phase 6) → PRIME Phase 7-10

PRIME is the orchestrator. Other skills are tools called at specific phases.

The 10 Phases

Phase 1 — CONCEPT

Question: What is this project and why should it exist?

Deliverables:

  • Logline — one sentence defining the core premise
  • Audience persona — who, pain points, behaviors
  • Differentiation — what this does that 3+ named competitors do NOT
  • Content mission — what the content does, for whom, to what end
  • Feasibility — time, budget, tools, can our pipeline produce this?
  • Format — platform, aspect ratio, duration, frequency, language

Exit Gate — ALL must be YES with written evidence before advancing.


Phase 2 — MARKET RESEARCH

Question: Who does this well, what works, what doesn't, where is the gap?

Methodology: Use the Viral Content Reverse-Engineering approach:

  • Collect: Outlier Score (views / channel median), View-to-Subscriber Ratio, 48h velocity, engagement ratios
  • Tools: Viral Finder (free) + OutlierKit ($9/mo) + Social Blade (free)
  • Distinguish content-driven virality (gradual ramp) from external-factor (spike + decay)
  • Minimum: 8-12 outlier videos from 3+ channels before conclusions

Deliverables:

  • Competitive audit — 10+ channels with metrics
  • Viral anatomy — 5-10 top pieces deconstructed second-by-second
  • Pattern identification — visual, audio, structural, title, timing patterns
  • Niche "invisible rules" — unwritten standards viewers expect
  • Content gaps — 3+ gaps between demand and supply quality
  • Revenue analysis — RPM, monetization ceiling, additional streams
  • Niche Intelligence Database started (Airtable: Videos + Channels + Patterns tables)

Exit Gate — ALL must be YES with written evidence before advancing.


Phase 3 — CONTENT STRATEGY

Question: Where do ideas come from, how is content planned, what is the editorial logic?

Deliverables:

  • Story sourcing system — documented methodology:
    • Trend mining (Google Trends, YouTube trending)
    • Competitor outlier detection (10x+ above channel avg)
    • Audience signals (bookmark comments = evergreen, tag comments = share trigger)
    • Arc requirements (narrative needs, not just trends)
    • Format transplantation (proven format from adjacent niche)
  • Content pillars — 3-5 thematic pillars every piece maps to
  • Season/series structure (if narrative) — arc, episode count, progression
  • Content brief template — standard format for one piece
  • Content calendar — 90 days populated with topics, formats, pillar assignments
  • Idea validation method:
    • Hook test: 4/5 people curious after first 3 seconds
    • Outlier precedent: 3+ videos with same mechanic are outliers
    • Emotion audit: specific emotion at seconds 0-3 and 15-22
    • Share test: "who sends this to whom and why" in one sentence

Exit Gate — ALL must be YES with written evidence before advancing.


Phase 4 — CREATIVE DEVELOPMENT

Question: What does this brand look and sound like? Who are the characters?

REQUIRED SUB-SKILL: Use superpowers:brainstorming here.

Deliverables:

  • Channel/brand identity — name, logo, banner, color palette (hex), typography
  • Tone of voice guide — formal/informal, serious/humorous, with do/don't examples
  • Production bible (biblia) — master document encoding ALL creative rules:
    • Identity, creative direction, visual production, audio production
    • SEO, what NOT to do, productionPackage JSON format, pipeline summary
  • Character design (if applicable):
    • cast.json with all recurring characters
    • Character sheet reference images (seed images, never regenerated)
    • Stable prompt strings per character (40-60 words, never modified)
  • Season arc document (if narrative) — episode-by-episode outline

Exit Gate — ALL must be YES with written evidence before advancing.


Phase 5 — PRODUCTION DESIGN

Question: What are the exact templates, standards, and specs that make content reproducible?

REQUIRED SUB-SKILL: Use superpowers:writing-plans here.

Deliverables:

  • Prompt library — versioned, modular (70% subject, 20% environment, 10% style)
  • Quality rubric — pass/revision/reject criteria per format
  • SOP for each production stage — brief to publish
  • Implementation plan — task-by-task build plan
  • Technical specs — resolution, FPS, audio format, loudness, bitrate

Exit Gate — ALL must be YES with written evidence before advancing.


Phase 6 — TECHNOLOGY & INFRASTRUCTURE

Question: Does the pipeline work? Can it produce content automatically?

REQUIRED SUB-SKILL: Use superpowers:executing-plans here.

Deliverables:

  • Pipeline code — NicheModule implemented, compiles, registered
  • API integrations — all external APIs tested individually
  • Automation workflows — n8n (Scout, Idea Generator, Creative Director)
  • Data infrastructure — Airtable, Drive folders, Telegram
  • Scheduling — overnight slots configured, no conflicts
  • End-to-end test — 1 video from Airtable trigger to Drive upload
  • Monitoring — Telegram alerts, daily summary, error isolation

Exit Gate — ALL must be YES with written evidence before advancing.


Phase 7 — TESTING & QA (PILOT)

Question: Does the output match the vision? Good enough to publish?

REQUIRED SUB-SKILL: Use superpowers:verification-before-completion here.

Deliverables:

  • Pilot batch — 4-6 videos covering full range of content types
  • Quality review — every pilot scored against rubric
  • Character consistency check (if applicable)
  • Audio quality check — listen to every output
  • Audience feedback — 3-5 target audience members review
  • Failure classification — creative / production / system
  • Bible and SOPs updated with pilot learnings

Exit Gate — ALL must be YES with written evidence before advancing.


Phase 8 — LAUNCH

Question: Is the first content live and performing?

Deliverables:

  • Success definition — metric + threshold + timeframe
  • Launch batch — 3-5 pieces published simultaneously
  • Platform setup — profile, description, links, playlists
  • Cross-platform distribution — YouTube + TikTok + Instagram + Facebook
  • Cross-network promotion — links between channels
  • Analytics baseline — Day 1 and Week 1 snapshots

Exit Gate — ALL must be YES with written evidence before advancing.


Phase 9 — OPTIMIZATION

Question: What's working, what's not, how do we improve?

Deliverables:

  • Weekly metrics review — KPIs every 7 days
  • 30-day analysis — top/bottom performers identified
  • A/B testing — hierarchy: hooks > titles > thumbnails > length > posting time
  • Underperformance diagnostic for every failing video:
    1. Did people click? (no = packaging)
    2. Past 3 seconds? (no = hook)
    3. Through middle? (no = pacing)
    4. Reach end? (no = perceived ending)
    5. Mechanic fire? (no = core mechanic)
  • 3+ evidence-based changes to strategy
  • Bible and calendar updated

Exit Gate — 60+ days of data, documented changes, sustainable operation.


Phase 10 — SCALING

Question: How to grow without losing quality?

Deliverables:

  • Scaling roadmap — volume, format expansion, platform expansion
  • Long-form bridge (if applicable) — Shorts to long-form pipeline
  • Content repurposing pipeline — each piece generates derivatives
  • 90-day sprint cycles — build / scale / optimize

Continuous Gate — quarterly review of quality, growth, sustainability.

Tracking Progress

At any point, report status as:

[PROJECT NAME] — PRIME Status
Phase 1 (Concept):     ✅ / ❌ / ⚠️ Partial
Phase 2 (Research):     ✅ / ❌ / ⚠️ Partial
Phase 3 (Strategy):     ✅ / ❌ / ⚠️ Partial
Phase 4 (Creative):     ✅ / ❌ / ⚠️ Partial
Phase 5 (Design):       ✅ / ❌ / ⚠️ Partial
Phase 6 (Technology):   ✅ / ❌ / ⚠️ Partial
Phase 7 (Pilot):        ✅ / ❌ / ⚠️ Partial
Phase 8 (Launch):       ✅ / ❌ / ⚠️ Partial
Phase 9 (Optimize):     ✅ / ❌ / ⚠️ Partial
Phase 10 (Scale):       ✅ / ❌ / ⚠️ Partial

Reference Documents

For detailed methodology at each phase:

  • Phase 2: docs/research/The Complete Viral Content Reverse-Engineering Methodology.md
  • Phases 1-7: docs/research/Universal Digital Content Production Blueprint.md
  • Phases 4-6: docs/research/The Unified AI Video Channel Production Framework.md
  • Full schema: docs/project-execution-schema.md

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.