tune-anchorslisted

Sweep anchor selections over a writing-voice corpus and report which tag query produces the best register outcome. Runs match-voice then tighten-style over (article × arm) combinations, ranks arms on the four-axis register composite, and optionally verifies the top candidates with an external detector. Triggers: tune anchors, sweep anchors, which anchors should I use, calibrate writing-voice, onboard corpus.
petar-djukic/writing-skills · ★ 4 · Data & Documents · score 72

Install: claude install-skill petar-djukic/writing-skills

# tune-anchors ## The objective Given a `writing-voice/` corpus and one or more target articles, answer: **which anchor-selection rule produces rewrites closest to the author's own register?** A rule is an *arm* — a set of filters over the manifest: `role=venue-voice`, `tags~clipped`, `pre_ai=true`. Different arms produce different anchor sets, and the model copies the register of whatever it sees. The right arm is the one whose anchors give the model a register worth copying. ## The pipeline Each full trial runs two model passes before measurement: 1. **match-voice** (`drive.py`) — voice rewrite using the arm's anchors 2. **tighten-style** (`tighten.py`) — remove AI-register artifacts (passive stacks, nominalizations, filler) using the same model family shown transformation pairs, not rules 3. **measure** — register markers on the tightened output The ranking reflects the final output quality, not the raw voice draft. Tightening is what removes the AI sound; ranking without it would penalize arms whose raw drafts carry fixable markers and reward arms whose markers survive tightening unchanged. `--no-tighten` skips step 2 and measures the raw voice draft instead. Use it to isolate the voice effect or to compare the tighten delta across arms. ## When to run - **Writing-voice onboarding.** After the manifest exists and before the first real rewrite. Output is the `--anchor-tags` query to use thereafter. - **After the corpus grows.** Pool sizes change, and guid