From thinking-frameworks-skills
Assigns Substack draft or published posts to the best-fitting section based on content and section promises in section-map.md. Used for editorial workflows to load voice overlays or batch classify.
How this skill is triggered — by the user, by Claude, or both
Slash command
/thinking-frameworks-skills:classify-post-to-sectionThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
```
Per post (draft or published):
- [ ] Step 1: Read post body (not just title)
- [ ] Step 2: Read section-map.md for all current section promises
- [ ] Step 3: Score post fit against each section's promise (specific, testable, voice register)
- [ ] Step 4: If top score clearly above second → assign that section
- [ ] Step 5: If ambiguous between two sections → propose both; writer picks
- [ ] Step 6: If no section scores above threshold → assign `unassigned`
- [ ] Step 7: Return: {section_slug, confidence, rationale}
For each section, compute fit on:
topics frontmatter intersect with the section's typical topic distribution?Draft: "KV Cache as a library card catalog" — full body on KV cache mechanics, diagram-heavy, cites Vaswani et al. and Dao et al.
Current sections:
kalshi-log: scoreboard-required, prediction markets / IPL. Promise match: low. Score: 1/5.agent-workshop: mechanism + architecture, code-fence-welcome. Promise match: high. Score: 5/5.Output: {section_slug: agent-workshop, confidence: high, rationale: "mechanism post with explicit paper citations; matches Agent Workshop register and promise"}.
section: X in frontmatter, respect it — this skill only proposes when frontmatter is missing or unassigned.npx claudepluginhub lyndonkl/claude --plugin thinking-frameworks-skillsChecks if posts in a substacker section still fit that section's promise. Flags drift at three levels (acceptable, borderline, genuine) and recommends promise rewrites when multiple posts in one section violate.
Writes, optimizes, and grows Substack newsletters and web posts including ghostwriting with voice matching, algorithm optimization, Notes strategy, SEO, growth tactics, and monetization planning.
Splits long-form articles into web-optimized sections with subheadings, pull-quote suggestions, image placement markers, and reading-time estimates for CMS-ready content.