paper-reading-research-framing — Claude Code skill: paper framing
paper-reading-research-framing turns one stable research lineage into a framing package for a paper: close works, citation support per claim, positioning and a writing blueprint.
no license ★ 240 updated 2026-09-18
npx skills add franklee16/academic-research-skills --skill paper-reading-research-framing -a claude-codeWhat paper-reading-research-framing does
The skill takes a single research lineage, not a pile of drafts, and turns it into a framing package for a paper. What you feed it is one leaf source: a reviewed experiment design, a committed method design, or a research question card. It creates a project folder at {workspace-root}/research-framing/{project-slug}/ (root defaults to workspace) and fills it with source_research.md, close_works.md, claim_support_bank.md, positioning_matrix.md, story_spine.md and writing_blueprint.md.
The core is comparison against 3–8 close works: their problem, method, setting and evaluation, and how yours differs. A second pass checks which paper claims the citations can carry and which need weaker wording. Two gates sit in the middle, one for picking close works and one for story and claim safety.
You get either a full Research Framing Package or a Provisional Framing Sketch. The sketch appears when the leaf source is not a reviewed experiment design, and then results stay planned rather than reported. Choosing a method, baselines or metrics is outside what this skill does.
What it helps with
- turn one research lineage into a paper pitch
- pick 3–8 close works and state the difference from each
- check which claims the citations actually support
- draft an introduction and related-work outline
- keep planned evidence separate from reported results
Example prompts for the agent
- Build a framing package from
workspace/experiment-design.mdand pick the close works it needs. - Go through the introduction claims and tell me which ones the citations do not support.
- Sketch the positioning from this research question card and flag where experiment evidence is still missing.
How to install paper-reading-research-framing
1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill paper-reading-research-framing -a claude-code2Manually: put the files in place
~/.claude/skills/08-research-framing/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/08-research-framing/
1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill paper-reading-research-framing -a codex2Manually: put the files in place
.agents/skills/08-research-framing/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill paper-reading-research-framing -a cursor2Manually: put the files in place
.cursor/skills/08-research-framing/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install franklee16/academic-research-skills lit-review/Paper-Reading-Skills-main/skills/08-research-framing2Manually: put the files in place
.github/skills/08-research-framing/SKILL.md1Skills CLI (Vercel)
npx skills add franklee16/academic-research-skills --skill paper-reading-research-framing -a gemini-cli2Platform CLI
gemini skills install https://github.com/franklee16/academic-research-skills --path lit-review/Paper-Reading-Skills-main/skills/08-research-framingManually: put the files in place
.devin/skills/08-research-framing/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-reading-research-framing
How to install paper-reading-research-framing in Claude Code?
Run npx skills add franklee16/academic-research-skills --skill 08-research-framing -a claude-code, and the skill file lands at ~/.claude/skills/08-research-framing/SKILL.md.
What does paper-reading-research-framing do?
It turns one stable research lineage into a paper-facing package: close works, a check of what each claim can cite, positioning, and an outline for the introduction and contributions. It does not design experiments or choose a method.
When do I get a package and when only a sketch?
A full package requires a reviewed experiment design as the leaf source, with claims, evidence routes and scope spelled out. Anything upstream of that produces a provisional sketch where results are marked as planned.