analyze-paper
Skill9 starsChanged 2 months agoApache-2.0View on GitHub
This skill extracts claims, models, proofs, and limitations from a research paper. You use it to analyze or summarize an academic paper.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- nh-tu-z/my-agent-skilltheir own version, first seen Nov 8, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- nh-tu-z/my-agent-skill★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository10 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- brainstormSkill
- clarify-goalSkill
- critiqueSkill
- formalize-problemSkill
- investigateSkill
- reaperSkill
- review-literatureSkill
- search-paperSkill
- write-paperSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 73
If PDF reading fails, you try individual pages with pages.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Analyze Paper
You must read and apply the language rules before you write any output.
Usage
You invoke this skill by name with a paper path. Slash-command hosts use /analyze-paper <args>.
analyze-paper path/to/paper.pdf
analyze-paper reaper-workspace/papers/2024-1234.pdf --goal "post-quantum threshold signatures" --output reaper-workspace/papers/2024-1234-notes.mdThe first argument without a flag gives the paper path.
--output <path>sets the output path. The default path isreaper-workspace/notes/paper-summary.md.--goal "<text>"supplies the research goal. This flag requires a Relevance section.
Instructions
1. Read the Paper
You read the paper with the host's file tool. You extract text first if the host cannot read PDFs. You follow ../reaper/references/paper-analysis.md:
- Pass 1: You read the abstract, introduction, conclusion, and theorem statements. You identify the main claims.
- Pass 2: You read the protocol, proof outlines, and figures. You identify the technical approach.
- Pass 3: You read the full proofs, appendices, and security reductions. You check each logical step.
With --goal, you use Pass 1 for all papers. You also use Pass 2 for papers with medium relevance. You use all three passes for papers with high relevance.
2. Extract Information
You copy formal claims verbatim. You record differences between the paper's claims and its proofs.
3. Write Output
You write the summary to the default path or the --output path.
The title is # Paper Summary: [Paper Title].
You use these level-two sections:
| Section | Content |
| Metadata | You copy Title, Authors, Venue/Year, Paper ID, and Link. |
| Problem Statement | You explain the problem and its importance. |
| System Model | You give concrete answers for every applicable dimension in ../reaper/references/model.md. |
| Construction Overview | You describe the protocol, technique, and components. |
| Key Results | You copy each theorem verbatim with its number, model, and proof technique. |
| Proof Technique | You explain lemmas, reductions, and steps that use the corruption threshold or network model. |
| Complexity Claims | You state communication, round, and computation costs. |
| Strengths | You assess novelty, methods, proofs, evaluation, and clarity. You label each strength major or minor. |
| Weaknesses | You identify incorrect methods, missing proofs, unsupported claims, unfair comparisons, and unclear text. You label each weakness major, minor, or fatal. |
| Key Definitions and Notation | You explain unusual notation and definitions that the proofs use. |
| Red Flags | You state specific concerns from ../reaper/references/paper-analysis.md, or you state that you found none. |
| Relevance | You include this section only with --goal. You explain each applicable tag in one sentence. |
Relevance tags are problem definition, formalization, solution technique, negative result, literature/context, and writing model.
You adjust section length to the paper's content. You omit sections that do not apply, such as absent complexity claims.
Quality Criteria
- You cover every applicable model dimension: network, adversary, trust, communication, and cryptography. You omit System Model only when unnecessary.
- The summary explains the claims and approach without the source paper.
- If PDF reading fails, you try individual pages with
pages. If reading still fails, you report an error without an invented summary.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.