Skip to main content
skill CX Codex

ccf-paper-to-exemplar — Claude Code skill: what it does

Turns the paper PDFs you feed it into writing exemplar cards and files them into your own library. Useful when you need real samples of structure and phrasing for a draft.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add mikubaka88/CCFA-Skills --skill ccf-paper-to-exemplar -a claude-code
Open on GitHub

What ccf-paper-to-exemplar does

The skill reads a paper and writes down how it works: the story it tells, what the abstract and introduction do, how the method and evidence sections move, which citation habits come back. That becomes a card with sections for Story Pattern, Abstract Moves, Introduction Moves, Method Moves, Evidence Moves, Citation Patterns, Reusable Techniques and Do-Not-Copy Boundary. Finished cards are registered in a personal library you can pull from later.

Extraction runs through scripts/convert.py with --output-dir, --full-text and --full-text-dir spelled out. That step wants Python with pymupdf. The converter writes only missing skeletons and leaves an existing card alone when the source version matches, so a rerun will not wipe your edits. Work files default to ccfa-workfiles/exemplars/<source-id>/, with raw text under cache/, away from the card library.

Attribution stays with the source. Argument structure and some language conventions carry over, scientific claims do not, and a Do-Not-Copy Boundary section marks that line. Read it before leaning on a card in your own draft: an empty extraction is not the same as a finished analysis.

What it helps with

  • turn a paper PDF into a writing exemplar card
  • build a personal library of sample papers
  • refresh one section of an existing card in place
  • hand the right writing moves to the drafting skill
  • check where copying a paper's claims stops

Example prompts for the agent

  • Turn papers/main.pdf into an exemplar card under ccfa-workfiles/exemplars/main/
  • Register this card in ../ccf-paper-writer/references/exemplars/index.md without duplicating entries
  • Update Reusable Techniques in the existing card, same source version

How to install ccf-paper-to-exemplar

1Skills CLI (Vercel)

npx skills add mikubaka88/CCFA-Skills --skill ccf-paper-to-exemplar -a claude-code

2Manually: put the files in place

~/.claude/skills/ccf-paper-to-exemplar/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/ccf-paper-to-exemplar/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about ccf-paper-to-exemplar

How do I install ccf-paper-to-exemplar in Claude Code?

Run npx skills add mikubaka88/CCFA-Skills --skill ccf-paper-to-exemplar -a claude-code. The skill lands at ~/.claude/skills/ccf-paper-to-exemplar/SKILL.md.

What does ccf-paper-to-exemplar do?

It reads a paper PDF section by section and fills a card with its writing moves: story, abstract, introduction, method, evidence, citations, reusable techniques. On request it also registers the card in your exemplar library.

What do I need for the PDF extraction to work?

Python with pymupdf, otherwise scripts/convert.py cannot pull the text out. Reusing a card that already exists for the same source needs no extraction.

Source text

SKILL.md · MIT
---
name: ccf-paper-to-exemplar
description: "Distill supplied paper PDFs into writing exemplar cards and register a requested personal library. Use for PDF to writing exemplar, 写作范例, and reusable paper-writing patterns. Preserve attribution; transfer argument structure and selected language conventions without copying scientific claims. Figure reconstruction, manuscript drafting, and review have separate owners."
metadata:
  ccf_skill_controls:
    handoff_question_mode: partial
    respect_session_denylists: true
    protect_idea_scope_in_writing: true
    private_material_safety: moderate
    shared_controls: ../ccf-common/references/
---

# Paper-to-Exemplar

## Family File Contract

Before file work, reuse canonical paths and one task/artifact working directory; default to project-root `ccfa-workfiles/<purpose>/<artifact-id>/`. Update in place, preserve inputs, and use UTF-8 with Chinese-text checks. Apply [artifact-contracts.md](../ccf-common/references/artifact-contracts.md) for naming, ownership, and task-created cleanup; handoffs inherit these paths.

## Collaboration Contract

Before specialist execution, read and apply [ccf-humanization](../ccf-humanization/SKILL.md) first, then [ccf-common](../ccf-common/SKILL.md). At every handoff, reuse their applicable active rules or refresh missing/changed ones. Both preflights are required even without prose; detailed editing, experiment, and maintenance modes run only when relevant.

Integrate necessary specialist contributions under one owner; reuse valid prerequisites and verify affected results. Follow the [cooperation routes](../ccf-common/references/routing.md) without duplicate reports.

## Scope And Shared Controls

Distill supplied paper PDFs into reusable writing patterns, preserving source attribution and scientific meaning. This skill owns exemplar cards; manuscript drafting belongs to `ccf-paper-writer`, visual reconstruction to `ccf-visual-composer`, and scientific review to `ccf-paper-reviewer`.

Follow `../ccf-common/references/handoff-modes.md`, `../ccf-common/references/task-modes.md`, and `../ccf-common/references/artifact-contracts.md`. Resolve existing card/library and cache paths before writing. Preserve no-new-files requests by editing a named existing card or returning the analysis in context. Conversion alone does not authorize global library/default changes.

## Workflow

1. Identify PDF paths, source versions, desired writing patterns, optional venue, and authorized destination. Reuse a completed card and extraction when they match the same source and requested scope; a new source version requires checking the affected analysis.
2. If extraction is needed, run `scripts/convert.py` with explicit `--output-dir`, `--full-text`, and `--full-text-dir` pointing to the chosen working cache. For a new project, the shared default is `ccfa-workfiles/exemplars/<source-id>/`, with extraction in `cache/`; preserve established paths. The converter preserves existing cards and creates only missing skeletons. `--full-text-dir` is optional; omitting it preserves the legacy output layout. Python with `pymupdf` is required for extraction, not for reusing a valid existing card.
3. Check extraction success and page markers. Analyze the relevant full source in coherent sections; use search and targeted ranges for a local card update. Inspect actual PDF pages when equations, figures, or reading order affect interpretation. Do not substitute an abstract-only analysis for a full-paper exemplar or treat empty extraction as complete.
4. Recover the source's story, abstract/introduction/method/evidence moves, citation patterns, and reusable techniques. Do not force a limitations ending. Use `../ccf-paper-writer/references/prose-quality-guardrails.md` to exclude defensive habits, artificial labels, formula dumping, unsupported hype, and mechanical prose from the advice.
5. Fill or revise the existing card in place using the section headings generated by `scripts/convert.py` and established cards: Story Pattern, Abstract Moves, Introduction Moves, Method Moves, Evidence Moves, Citation Patterns, Reusable Techniques, and Do-Not-Copy Boundary. Record title, venue/year when verified, source identity/version, and relevant page/section anchors. When language adaptation is requested, include the relevant punctuation, sentence rhythm, or transition conventions under Reusable Techniques with source anchors and their intended use. No new profile file or corpus quota is needed. Add Excluded Anti-Patterns only when the source supplies a material example. Keep writing moves separate from source claims and distinctive wording.
6. When library registration is requested, use `../ccf-paper-writer/references/exemplars/cards/` or the user's existing library and update `../ccf-paper-writer/references/exemplars/index.md` without duplicate entries. Preserve unrelated cards; a matching filename alone does not prove the same source. Keep raw extraction in the working cache, outside the reusable card library.
7. Update `../ccf-paper-writer/references/custom-format/default-user-format.md` only when default selection is requested. A venue-only request can select an existing relevant card through the index instead of extracting another PDF. Load `../ccf-paper-writer/references/venue-guides/index.md` only when venue-family classification is unclear.
8. Verify that completed cards have no `[ANALYZE]` or `[MANUAL]` placeholders, source references are accurate, and output/cache paths are orderly. Return the canonical card paths and the useful writing patterns; report actual registration/default changes only when performed. Do not echo the full PDF extraction or add generic next-action menus.

## Writer Handoff

Pass the selected card path, source identity, and relevant writing moves. The writer decides whether a style reference is needed for its task; a paragraph edit does not automatically load default or venue card bundles. For full manuscripts or explicit style adaptation, select from `../ccf-paper-writer/references/exemplars/index.md` and load only matching sections/cards. Preserve existing user defaults without treating them as mandatory reading for every writing request.

© mikubaka88 · MIT · mikubaka88/CCFA-Skills

Similar skills

Everything for Writing and docs →

From the same set5

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor