Format First

Decide WHAT SHAPE of content to make, and refuse the platform folklore that circulates as fact. Encodes a claim set traced to primary sources — Meta's own engineering writeup and Transparency Center system cards, Mosseri's actual posts, and the psychology literature behind "hooks" and repetition — separating what is documented, what is hypothesis, and what is unsupported. Ships a linter that GRADES each claim — refuted, contested, unverified, folklore — and errors only on the ones contradicted by a primary source that was actually loaded, because "no source found" is a statement about a search, not about the world. The core move is choosing a FORMAT (visual grammar x script structure) as the invariant and letting topic vary, rather than the reverse. Use when, PRE-PRODUCTION: (1) choosing what shape a piece should take before it is written, (2) deciding how often to repeat a format and when to rotate it, (3) verifying a platform/algorithm claim from a course, thread, or guru, (4) auditing a draft or strategy doc for recommender folklore and untraceable statistics. NOT FOR: producing the artifact (use content-creation), generating visuals (use content-engine), page markup (use seo-llmeo), running an engagement loop (use social-intelligence), or **anything about a finished piece's reach, length, or citation by AI answer engines — that is citable, including the short-post-vs-long-article decision.** Triggers on "what format should I use", "is the algorithm punishing me", "should I post daily", "does this algorithm claim hold up", "which format should I repeat", "3-second hook", "verify this platform claim".

mkdir -p ~/.claude/skills/format-first
curl -L https://claudskills.com/skills/format-first/SKILL.md \
  -o ~/.claude/skills/format-first/SKILL.md

@misc{format-first-2026,
  author    = {broomva},
  title     = {Format First [Claude Code skill]},
  year      = {2026},
  publisher = {ClaudSkills},
  url       = {https://claudskills.com/skills/format-first/}
}