paper-evidence-distiller — Claude Code skill: guide and install
paper-evidence-distiller extracts structured evidence cards from academic papers in PDF or Markdown format. Links every empirical claim directly to its source section or table.
no license ★ 240 updated 2026-09-18
npx skills add franklee16/academic-research-skills --skill paper-evidence-distiller -a claude-codeWhat paper-evidence-distiller does
The skill reads research papers provided as .pdf or .md files and generates structured .distilled.md notes. Instead of broad abstracts, it isolates sample parameters, timeframes, empirical specifications, and observed effect directions. Abstract text alone never counts as proof. Every recorded finding requires an explicit reference to a table, figure, or section in the source file.
When processing PDF documents, the tool calls scripts/pdf_to_markdown.py to prepare an intermediate Markdown copy. This helper automatically pulls pypdf and pdfplumber dependencies unless PAPER_EVIDENCE_DISTILLER_NO_INSTALL=1 is configured in the environment. Output files are placed in the distilled/ directory by default. Without an exact source anchor in the text, data is marked as unconfirmed rather than guessed.
What it helps with
- extract empirical evidence from research papers
- tie scientific findings to exact tables and sections
- convert paper PDFs into intermediate Markdown files
- build searchable notes for literature review chapters
Example prompts for the agent
- Create an evidence card for
papers/study.pdf - Extract methods and findings with anchors from
papers/analysis.md - Process all paper files in
papers/and save cards todistilled/
How to install paper-evidence-distiller
1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill paper-evidence-distiller -a claude-code2Manually: put the files in place
~/.claude/skills/paper-evidence-distiller/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/paper-evidence-distiller/
1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill paper-evidence-distiller -a codex2Manually: put the files in place
.agents/skills/paper-evidence-distiller/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill paper-evidence-distiller -a cursor2Manually: put the files in place
.cursor/skills/paper-evidence-distiller/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install franklee16/academic-research-skills lit-review/paper-evidence-distiller/SKILL.md2Manually: put the files in place
.github/skills/paper-evidence-distiller/SKILL.md1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill paper-evidence-distiller -a gemini-cli2Platform CLI
gemini skills install https://github.com/franklee16/academic-research-skills --path lit-review/paper-evidence-distillerManually: put the files in place
.devin/skills/paper-evidence-distiller/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about paper-evidence-distiller
How do I install paper-evidence-distiller in Claude Code?
Run npx skills add franklee16/academic-research-skills --skill paper-evidence-distiller -a claude-code in your terminal.
How does an evidence card differ from a regular summary?
It prioritizes verifiable facts such as sample metrics, variables, and effect directions over general prose. Every finding must point to a specific section, table, or figure from the original text.
What dependencies are needed to handle PDF files?
The skill uses scripts/pdf_to_markdown.py, which requires Python. It installs pypdf and pdfplumber automatically unless PAPER_EVIDENCE_DISTILLER_NO_INSTALL=1 is set.