# Writing Shape · dotknewt

> Use when the user has raw material — a pile of notes, a transcript, unstructured prose — and wants it shaped into a finished article one paragraph (or block) at a time, arguing through 2-3 candidate openings and format choices (prose vs. list, quote vs. paraphrase) with the user before each piece is appended to the article file, instead of getting a full draft dumped at once. Triggers on requests like "here's a braindump, turn it into an article", "help me turn this into a post", "shape this pile into something readable, paragraph by paragraph", or "build a writeup from these notes and argue the structure with me as we go" — even when they don't say "article" or "writing" explicitly. Not for open-ended brainstorming before raw material exists (see writing-fragments), a choose-your-own-adventure journey assembled beat by beat (see writing-beats), or line-editing an already-finished draft with no raw-material pile behind it (see edit-article). disable-model-invocation is intentional while this skill is still maturing in skills/in-progress/: the workflow hasn't been validated for reliable auto-triggering yet (see evals/evalqueries.json for the intended trigger scenarios, documented now so they're ready to test once this flag is revisited). It stays top-level rather than under metadata: (the base skill-spec's documented extension point for inert key-value data) because Claude Code reads this field from the top level to actually suppress auto-invocation; metadata is not consulted for behavior. This is the same convention used by sibling skills in this repo (loop-me, wizard, writing-fragments, writing-beats, claude-handoff, edit-article) — moving it under metadata would risk silently re-enabling auto-triggering for a skill not yet vetted for it.

## Facts
- Page: https://tashan.sh/capability/skill-dotknewt-writing-shape
- tashan id: skill:dotknewt/writing-shape
- Source: https://github.com/dotknewt/awesome-agency
- Type: skill
- Category: other
- tashan score: not scored (catalogued only — too little public evidence)
- Adoption: 9.0
- Upkeep: 96.0
- Freshness: 92.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: not yet graded
- Official: no

## Install

```sh
cp -r writing-shape ~/.claude/skills/
```

## Security audit
Not scanned. We audit npm-published capabilities; this one has no npm package we can resolve, or has not reached the queue. This is not a clean bill of health.

---
Measured 2026-09-12 by tashan (https://tashan.sh) from public evidence. Scorer s5.