Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Fill `outline/tables_index.md` from `outline/table_schema.md` + evidence packs (NO PROSE in cells; citation-backed rows). **Trigger**: table filler, fill tables, evidence-first tables, index tables, 表格填充, 索引表. **Use when**: table schema exists and evidence packs are ready; you want a compact, citation-backed index table to support later writing and Appendix table curation.
.claude/skills/willoscar-table-filler/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 0% | 0% |
Goal: produce outline/tables_index.md as an internal, citation-backed index.
This file is a planning/traceability artifact:
Reader-facing tables belong in outline/tables_appendix.md and are curated by appendix-table-writer.
Treat this as filling a table artifact from evidence packs:
outline/table_schema.md first (it defines what each table must answer).queries.md:draft_profile when present (course_paper requires one index table; survey / deep require two).outline/subsection_briefs.jsonl + outline/evidence_drafts.jsonl to fill the subsection map and pick in-scope citations.outline/anchor_sheet.jsonl to fill the anchor-fact table (prefer quant/eval hooks).citations/ref.bib.outline/table_schema.mdoutline/subsection_briefs.jsonloutline/evidence_drafts.jsonloutline/anchor_sheet.jsonlcitations/ref.biboutline/tables_index.mdoutline/tables_index.md must:
course_paper, or >=2 for survey / deep**Index Table 1. ...**#, ##, ###) unless you explicitly want them for internal readability[@BibKey]TODO, ..., …, scaffold comments)<br> sparingly)1) Subsection map (axes + representative works)
subsection_briefs.axes.2) Concrete anchors (benchmarks / numbers / caveats)
outline/anchor_sheet.jsonl (curated, citation-backed).evidence-selfloop / evidence-draft) OR narrow the schema.uv run python .codex/skills/table-filler/scripts/run.py --helpuv run python .codex/skills/table-filler/scripts/run.py --workspace <workspace>--workspace <workspace> (required)--unit-id <id> (optional; used only for runner bookkeeping)--inputs <schema;briefs;packs;anchors;bib> (optional; override inputs)--outputs <relpath> (optional; defaults to outline/tables_index.md)--checkpoint <C#> (optional; ignored by the bootstrapper)uv run python .codex/skills/table-filler/scripts/run.py --workspace <workspace>
uv run python .codex/skills/table-filler/scripts/run.py --workspace <workspace> --outputs outline/tables_index.md
Notes:
appendix-table-writer.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | fail→fail | 15,063 | 8,212 | -45% | 1 | 1 | 0% | 2,702 | 1,415 | -48% | 0 | 0 | — |
case-14 | fail→fail | 15,258 | 1,319 | -91% | 1 | 1 | 0% | 2,830 | 1,099 | -61% | 0 | 0 | — |
case-01 | fail→fail | 3,354 | 6,343 | +89% | 1 | 1 | 0% | 194 | 1,391 | +617% | 0 | 0 | — |
case-02 | fail→fail | 25,494 | 6,053 | -76% | 1 | 1 | 0% | 4,901 | 1,183 | -76% | 0 | 0 | — |
case-03 | fail→fail | 9,051 | 14,460 | +60% | 1 | 1 | 0% | 1,893 | 3,147 | +66% | 0 | 0 | — |
case-04 | fail→pass | 32,801 | 18,852 | -43% | 1 | 1 | 0% | 6,162 | 4,625 | -25% | 0 | 0 | — |
case-05 | fail→fail | 9,977 | 5,557 | -44% | 1 | 1 | 0% | 2,076 | 1,199 | -42% | 0 | 0 | — |
case-06 | pass→pass | 35,292 | 28,781 | -18% | 1 | 1 | 0% | 6,176 | 6,236 | +1% | 0 | 0 | — |
case-07 | fail→fail | 17,947 | 6,528 | -64% | 1 | 1 | 0% | 2,940 | 1,309 | -55% | 0 | 0 | — |
case-09 | pass→fail | 9,469 | 5,540 | -41% | 1 | 1 | 0% | 1,727 | 1,204 | -30% | 0 | 0 | — |
case-10 | fail→pass | 12,315 | 4,713 | -62% | 1 | 1 | 0% | 1,995 | 1,522 | -24% | 0 | 0 | — |
case-11 | pass→fail | 11,174 | 6,513 | -42% | 1 | 1 | 0% | 1,746 | 1,261 | -28% | 0 | 0 | — |
case-12 | pass→pass | 13,657 | 4,224 | -69% | 1 | 1 | 0% | 2,106 | 1,536 | -27% | 0 | 0 | — |
case-13 | pass→pass | 13,083 | 23,613 | +80% | 1 | 1 | 0% | 2,422 | 4,544 | +88% | 0 | 0 | — |
case-15 | fail→pass | 9,331 | 3,346 | -64% | 1 | 1 | 0% | 1,310 | 1,441 | +10% | 0 | 0 | — |
case-16 | fail→pass | 20,296 | 2,249 | -89% | 1 | 1 | 0% | 1,231 | 1,224 | -1% | 0 | 0 | — |
case-17 | fail→pass | 15,117 | 1,876 | -88% | 1 | 1 | 0% | 1,113 | 1,117 | +0% | 0 | 0 | — |
case-18 | fail→pass | 5,920 | 2,448 | -59% | 1 | 1 | 0% | 853 | 1,262 | +48% | 0 | 0 | — |
case-19 | fail→pass | 12,116 | 2,176 | -82% | 1 | 1 | 0% | 1,874 | 1,289 | -31% | 0 | 0 | — |
case-20 | fail→fail | 8,437 | 1,855 | -78% | 1 | 1 | 0% | 1,301 | 1,163 | -11% | 0 | 0 | — |
case-21 | pass→pass | 8,469 | 4,586 | -46% | 1 | 1 | 0% | 1,300 | 1,581 | +22% | 0 | 0 | — |
case-22 | fail→pass | 9,452 | 1,634 | -83% | 1 | 1 | 0% | 1,367 | 1,127 | -18% | 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 14 counted toward the lift figure. The other 8 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 +27 percentage points is the difference between those two pass rates over the 14 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
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.