YouTube Strategy

SkillMedia

'Builds a YouTube channel launch strategy for B2B SaaS operators. Outputs keyword demand analysis, competitor

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the YouTube Strategy skill

What this skill tells your AI

The instructions your AI receives, as published by matteotitta/genesys-skills in skills/primitives/social/youtube/youtube-strategy/SKILL.md and read by ahel’s review.

Generate a YouTube channel launch strategy for B2B SaaS operators. Output is a deck-style markdown document with keyword demand research, competitor segmentation, gap analysis, 6 video ideas, case-study / testimonial framework, and a Month-1 plan. Feeds into youtube-scripts for per-video production.

For full process, frameworks, and MCP integration → the premium reference.


Research Substrate (Exa)

Default: Exa per .claude/rules/exa-protocol.md.

Primary tools: web_search_exa for competitor channel + keyword research.

Tool surface during migration: prefer mcp__plugin_exa_exa__web_search_exa (after claude plugin i exa@claude-plugins-official); legacy mcp__exa__web_search_exa still mounted.

Citation: every claim uses [VERIFIED: exa_search, {url}, accessed {YYYY-MM-DD}] per .claude/rules/ontology.md.

Quality gate: ≥3 sources per major claim, ≥50% [VERIFIED], date filter for any "recent / latest" claim.


Claude Code Triggers

Invoke when user says:

  • "YouTube strategy for [company]"
  • "Plan my YouTube channel"
  • "What videos should I make on YouTube"
  • "YouTube channel strategy / plan / launch plan"
  • "Help me start a YouTube channel"
  • "What's the opportunity on YouTube"
  • "YouTube keyword research + video ideas"

Do NOT invoke when:

  • User wants individual YouTube script → use youtube-scripts
  • User wants LinkedIn video content → use linkedin-content
  • User wants multi-channel content strategy → use content-strategy
  • User wants content calendar / execution → use content-operations

Input Requirements

Client context (auto-loaded)

If working on a client project, client CLAUDE.md is auto-loaded. Pull positioning from knowledge/positioning.md and ICP from knowledge/icp-definition.md.

Required

InputDescriptionSource
Positioning statement1-2 sentence who-you-are + what-you-doknowledge/positioning.md or user
ICP definitionWho the channel targetsknowledge/icp-definition.md or user

Recommended (improves quality significantly)

InputHow it helps
Social engagement dataLinkedIn post reactions/comments per topic — fuels "proven demand" per idea
Client rosterCase-study candidate shortlist for BOFU videos
Content auditLets skill skip topics already covered

Optional

InputHow it helps
Niche hypothesisSharpens keyword seeding
Existing YouTube presenceAdjusts scope (launch vs relaunch vs refresh)
Competitor hypothesesSeed list for channel scan

Validation checklist

Before proceeding:

  • Positioning is 1-2 sentences with operator's specific angle (not generic)
  • ICP is named (title + company type), not "small businesses" or "B2B"
  • If LinkedIn data provided, it has post-level reactions/comments (not just topic themes)

If positioning or ICP is missing, ask before proceeding. Do not hallucinate.


Anti-Hallucination Guardrails

  1. No invented keyword volumes — always pull from Ahrefs MCP. If unavailable, mark [UNAVAILABLE: Ahrefs pull failed].
  2. No invented engagement data — "proven demand" per idea must cite real LinkedIn/Twitter/newsletter signal. If none, mark [UNAVAILABLE].
  3. No invented case-study metrics — pull from knowledge/positioning.md proof-points or user.
  4. No invented rankings — for Phase 3 gap analysis, mark [UNVERIFIED: manual check needed] if YouTube SERP not run.
  5. Flag gap claims with source — "nobody covers X" needs [INFERRED: from competitor scan on {date}].
  6. Client roster dependency — pull from client's knowledge/profile.md clients list or explicit user input.

Process, Output, Quality, and Auto-Update

TopicReference
5-phase process + frameworks + MCP integrationthe premium reference
Output template + render targets + iteration prompts + handoffthe premium reference
Pre-delivery checklist + worked example + anti-examples + quality gatethe premium reference
5-bucket competitor taxonomy with detection rulesthe premium reference
6-field video idea template with examplesthe premium reference
Month-1 mix rules + funnel outcomesthe premium reference
Ahrefs MCP calls + competition scoring methodthe premium reference
Worked example — Matteo's own YouTube strategyexamples/0426-matteo-youtube-strategy.md

Skill Relationships (per ontology.md)

depends_on:
  - company-context # required — pulls positioning, clients, services
  - icp-profile # required — defines channel audience
  - positioning # recommended — sharpens "your edge" per idea
feeds_into:
  - youtube-scripts # per-video script production
  - content-operations # add Month-1 plan to weekly calendar
  - case-study # BOFU case-study production
enhances:
  - content-strategy # if multi-channel strategy exists, refines YouTube lane

Knowledge type produced: youtube-strategy (Level 1 — Strategy, review gate 2).


Final ship gate

Run /premortem --output before ship. See /premortem skill for the 5 execution domains (will-it-resonate / will-it-convert / will-it-stay-on-brand / will-stakeholder-push-back / will-it-degrade-over-time) and output template.

Trivial-case escape: ## Premortem\nNo failure modes — trivial change satisfies the contract for genuinely trivial outputs.


Signals

GitHub stars
36
Forks
14
Last commit
Jul 2026
Advanced
Catalog kind
skill
Gateway key
youtube-strategy-matteotitta
Source
github.com/matteotitta/genesys-skills