本文へ移動
cccskills
無料GitHub で公開

table-filler

Fill `outline/tables_index.md` from `outline/table_schema.md` + evidence packs (NO PROSE in cells; citation-backed rows).

インストール方法を見る

含まれるファイル(2)

  • SKILL.md3.9 KB
  • scripts/run.py10.5 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Table Filler (index tables, evidence-first)

Triggers & routing

  • 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.

Goal: produce outline/tables_index.md as an internal, citation-backed index.

This file is a planning/traceability artifact:

  • it is useful for debugging coverage and for quickly seeing what evidence exists
  • it should NOT be pasted into the paper verbatim

Reader-facing tables belong in outline/tables_appendix.md and are curated by appendix-table-writer.

Default mode: semantic (LLM-first)

Treat this as filling a table artifact from evidence packs:

  • restate evidence pack content in compact cells
  • attach citations in the row (avoid cite-dump paragraphs)

Workflow (explicit inputs)

  • Read outline/table_schema.md first (it defines what each table must answer).
  • Read queries.md:draft_profile when present (course_paper requires one index table; survey / deep require two).
  • Use outline/subsection_briefs.jsonl + outline/evidence_drafts.jsonl to fill the subsection map and pick in-scope citations.
  • Use outline/anchor_sheet.jsonl to fill the anchor-fact table (prefer quant/eval hooks).
  • Validate cite keys against citations/ref.bib.

Inputs

  • outline/table_schema.md
  • outline/subsection_briefs.jsonl
  • outline/evidence_drafts.jsonl
  • outline/anchor_sheet.jsonl
  • citations/ref.bib

Output

  • outline/tables_index.md

Output format contract

outline/tables_index.md must:

  • contain >=1 Markdown table for course_paper, or >=2 for survey / deep
  • use a caption line before each table, e.g. **Index Table 1. ...**
  • contain no Markdown headings (#, ##, ###) unless you explicitly want them for internal readability
  • include citations in rows using [@BibKey]
  • avoid placeholders (TODO, ..., …, scaffold comments)
  • avoid paragraph cells (short phrases; use <br> sparingly)

What to fill (recommended defaults)

  1. Subsection map (axes + representative works)
  • Axes come from subsection_briefs.axes.
  • Representative works come from citations in the evidence pack.
  1. Concrete anchors (benchmarks / numbers / caveats)
  • Anchor facts come from outline/anchor_sheet.jsonl (curated, citation-backed).
  • Representative works come from citations in the anchor sheet (or the evidence pack as fallback).

Common failure modes

  • Cells become long prose

    • Fix: compress into short phrases; move narrative explanation into prose sections.
  • Rows cannot be filled without guessing

    • Fix: treat as an evidence gap (route to evidence-selfloop / evidence-draft) OR narrow the schema.

Script (optional bootstrap)

Quick Start

  • uv run python .codex/skills/table-filler/scripts/run.py --help
  • uv run python .codex/skills/table-filler/scripts/run.py --workspace <workspace>

All Options

  • --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)

Examples

  • Bootstrap index tables (default output path):

    uv run python .codex/skills/table-filler/scripts/run.py --workspace <workspace>

  • Write to a custom index file (rare):

    uv run python .codex/skills/table-filler/scripts/run.py --workspace <workspace> --outputs outline/tables_index.md

Notes:

  • The script is deterministic bootstrap; treat the result as an internal index artifact.
  • Curate reader-facing Appendix tables separately via appendix-table-writer.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Internal template for creating or refactoring a skill into the repository's reference-first shape. **Trigger**: reference-first template, blueprint skill, create a reusable skill, refactor a script-heavy skill. **Use when**: you need a lean `SKILL.md`, explicit `references/`, machine-readable `assets/`, and a minimal deterministic `run.py`.

日本語の概要は準備中です。原文の説明を表示しています。

WILLOSCAR/research-units-pipeline-skills5132026年10月5日 更新

Download a small corpus of open-access arXiv survey/review PDFs about agentic systems and extract text for style learning.

日本語の概要は準備中です。原文の説明を表示しています。

WILLOSCAR/research-units-pipeline-skills5132026年10月5日 更新

Extract per-subsection “anchor facts” (NO PROSE) from evidence packs so the writer is forced to include concrete numbers/benchmarks/limitations instead of generic summaries.

日本語の概要は準備中です。原文の説明を表示しています。

WILLOSCAR/research-units-pipeline-skills5132026年10月5日 更新

Curate reader-facing survey tables for the Appendix (clean layout + high information density), using only in-scope evidence and existing citation keys.

日本語の概要は準備中です。原文の説明を表示しています。

WILLOSCAR/research-units-pipeline-skills5132026年10月5日 更新

Build a final, machine-readable argument snapshot from current H3 sections and refresh section fingerprints before merge.

日本語の概要は準備中です。原文の説明を表示しています。

WILLOSCAR/research-units-pipeline-skills5132026年10月5日 更新

Audit one research Workspace for declared Unit outputs and Pipeline target Artifacts, writing `output/CONTRACT_REPORT.md`; use for mid-Run coverage snapshots or final delivery completeness, not deep provenance integrity.

日本語の概要は準備中です。原文の説明を表示しています。

WILLOSCAR/research-units-pipeline-skills5132026年10月5日 更新

WILLOSCAR のスキルをすべて見る

このスキルの問題を報告する