Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Bind papers to chapter-level sections first, writing `outline/section_bindings.jsonl` and `outline/section_binding_report.md`. **Trigger**: section bindings, chapter bindings, section-first binding, 章节绑定, 章级绑定. **Use when**: survey structure should measure chapter saturation before stable H3 decomposition.
.claude/skills/willoscar-section-bindings/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -39% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -46% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -16% | 0% |
Always read:
references/overview.mdUse scripts/run.py only for deterministic binding and report materialization:
papers/core_set.csv as the authoritative paper pooloutline/chapter_skeleton.yml as the chapter contractpapers/papers_dedup.jsonl when available for richer metadatapapers/core_set.csvoutline/chapter_skeleton.ymlpapers/papers_dedup.jsonloutline/section_bindings.jsonloutline/section_binding_report.mdassets/output_contract.jsonuv run python .codex/skills/section-bindings/scripts/run.py --workspace <workspace>--workspace <dir>: workspace containing the core set and chapter skeleton--unit-id <id>: optional harness metadata--inputs <semicolon-separated>: optional override from UNITS.csv--outputs <semicolon-separated>: optional output override; defaults are outline/section_bindings.jsonl and outline/section_binding_report.md--checkpoint <C*>: optional harness metadatauv run python .codex/skills/section-bindings/scripts/run.py --workspace <workspace> --inputs 'papers/core_set.csv;outline/chapter_skeleton.yml;papers/papers_dedup.jsonl' --outputs 'outline/section_bindings.jsonl;outline/section_binding_report.md'| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | fail→pass | 8,354 | 2,172 | -74% | 1 | 1 | 0% | 1,334 | 692 | -48% | 0 | 0 | — |
case-16 | fail→pass | 16,466 | 7,265 | -56% | 1 | 1 | 0% | 2,544 | 1,557 | -39% | 0 | 0 | — |
case-01 | fail→fail | 7,050 | 5,457 | -23% | 1 | 1 | 0% | 487 | 676 | +39% | 0 | 0 | — |
case-02 | fail→fail | 14,494 | 5,638 | -61% | 1 | 1 | 0% | 2,610 | 621 | -76% | 0 | 0 | — |
case-03 | fail→fail | 9,748 | 4,249 | -56% | 1 | 1 | 0% | 1,712 | 510 | -70% | 0 | 0 | — |
case-04 | fail→pass | 9,340 | 2,864 | -69% | 1 | 1 | 0% | 1,465 | 796 | -46% | 0 | 0 | — |
case-06 | fail→pass | 12,499 | 2,337 | -81% | 1 | 1 | 0% | 1,750 | 829 | -53% | 0 | 0 | — |
case-07 | fail→pass | 5,461 | 1,682 | -69% | 1 | 1 | 0% | 754 | 637 | -16% | 0 | 0 | — |
case-08 | fail→pass | 8,626 | 2,101 | -76% | 1 | 1 | 0% | 1,364 | 726 | -47% | 0 | 0 | — |
case-09 | fail→pass | 5,326 | 1,960 | -63% | 1 | 1 | 0% | 774 | 682 | -12% | 0 | 0 | — |
case-15 | pass→pass | 9,995 | 1,779 | -82% | 1 | 1 | 0% | 1,448 | 667 | -54% | 0 | 0 | — |
case-10 | fail→pass | 5,839 | 1,837 | -69% | 1 | 1 | 0% | 837 | 624 | -25% | 0 | 0 | — |
case-11 | fail→pass | 4,116 | 1,527 | -63% | 1 | 1 | 0% | 558 | 575 | +3% | 0 | 0 | — |
case-12 | pass→pass | 11,888 | 2,596 | -78% | 1 | 1 | 0% | 2,032 | 854 | -58% | 0 | 0 | — |
case-13 | fail→pass | 26,604 | 1,528 | -94% | 1 | 1 | 0% | 1,877 | 643 | -66% | 0 | 0 | — |
case-14 | fail→pass | 10,461 | 2,233 | -79% | 1 | 1 | 0% | 1,658 | 809 | -51% | 0 | 0 | — |
case-17 | fail→pass | 8,139 | 2,109 | -74% | 1 | 1 | 0% | 1,376 | 708 | -49% | 0 | 0 | — |
case-18 | pass→pass | 11,394 | 1,158 | -90% | 1 | 1 | 0% | 893 | 545 | -39% | 0 | 0 | — |
case-19 | fail→pass | 8,944 | 1,816 | -80% | 1 | 1 | 0% | 1,362 | 595 | -56% | 0 | 0 | — |
case-20 | fail→fail | 14,854 | 5,041 | -66% | 1 | 1 | 0% | 2,341 | 572 | -76% | 0 | 0 | — |
case-21 | fail→fail | 17,661 | 4,576 | -74% | 1 | 1 | 0% | 1,906 | 649 | -66% | 0 | 0 | — |
case-22 | fail→fail | 18,927 | 7,389 | -61% | 1 | 1 | 0% | 4,110 | 921 | -78% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 16 counted toward the lift figure. The other 6 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +59 percentage points is the difference between those two pass rates over the 16 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.