cc-content-atomizelisted

@../\_shared/storytelling-frameworks.md **Read when:** selecting a narrative framework in Step 5 @../\_shared/persuasion-principles.md **Read when:** selecting persuasion principles in Step 6 # Distribution Engine (Atomization) Skill Takes one core message — either a `brief.md` campaign handoff or a manually described idea — and produces properly-adapted drafts for every named format in a single run, keeping the core claim and proof points identical across all of them while letting structure, length, and tone vary per format's own conventions. **Key principle (from research):** fix the facts, vary the frame. The specific claim, statistics, named proof points, product/entity names, and quoted figures must stay **byte-identical** across every atomized piece; structure, length, hook, opening, tone, CTA, and formatting must be **rewritten per platform** to match its native conventions. This skill enforces that separation to prevent "copy-paste repurposing" — a truncated blog dumped onto LinkedIn — which audiences and algorithms both punish. ## Step 0: Recall learnings If `.claude/learnings.md` exists, read it silently, `[cc-content:*]` tags plus any cross-plugin entries relevant to distribution/repurposing. Never announce. ## Step 1: Load context Read the context table from all loaded CLAUDE.md files: ```bash grep -A 200 '## Context files' CLAUDE.md 2>/dev/null || echo "(no context table)" ``` CLAUDE.md files may exist at multiple hierarchy levels (workspace root, projec