Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add chgagne/claude-skills-research --skill comparing-papersgit clone --depth 1 https://github.com/chgagne/claude-skills-researchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/chgagne/claude-skills-research/comparing-papers)<a href="https://agentmods.dev/skills/chgagne/claude-skills-research/comparing-papers"><img src="https://agentmods.dev/badge/skills/chgagne/claude-skills-research/comparing-papers/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/chgagne/claude-skills-research/comparing-papers"><img src="https://agentmods.dev/badge/skills/chgagne/claude-skills-research/comparing-papers.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00091 | $0.01316 |
| Opus 5 | $0.00046 | $0.00658 |
| Sonnet 5 | $0.00018 | $0.00263 |
| Haiku 4.5 | $0.00009 | $0.00132 |
Grade A, and why
comparing-papers scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Comparing Papers
Overview
Puts a draft and its closest prior work side by side on the axes reviewers attack, with every cell traceable to a quoted sentence and the section it came from.
The tool gathers evidence; you draw the conclusion. It computes a note on exactly two axes — a training-scale ratio and a seed count — because those are arithmetic. On every other axis it shows two passages and stops. "These protocols differ" reads like a finding but is a judgement, and a tool that makes it will eventually make it wrongly in a document you sign your name to.
This automates the check reviewing-paper-sources/reference/claim-audit.md calls
"baseline training scale vs its published scale", which otherwise means reading two
appendices and doing a multiplication.
Run it
python3 ~/.claude/skills/comparing-papers/assets/run-compare.py . \
--against "SNIP: Bridging Mathematical Symbolic and Numeric Realms" \
--out review-assets/
# take the shortlist straight from a gap sweep
python3 ~/.claude/skills/comparing-papers/assets/run-compare.py . \
--from-candidates review-assets/candidates.json --grade THREAT
Run by absolute path from the paper directory. Stdlib only. --against accepts a title, a
DOI or an arXiv id and repeats; --limit caps how many papers are fetched (default 5).
Writes paper-comparison-<date>.md and comparison.json. Exit code 2 means some paper
was reachable only as an abstract.
The acquisition ladder
1. arXiv e-print source LaTeX, read with stdlib tarfile/gzip
2. arXiv PDF via pdftotext, only if that binary is installed
3. open-access PDF via OpenAlex oa_url, same caveat
4. abstract degraded
LaTeX source leads for a reason. Training scale, seed counts and compute budgets live in appendices, and appendices are what PDF extraction mangles. Rung 1 needs no PDF parsing at all and keeps section headings exact. On the SNIP paper it yields 66 sections including "Pre-training Data Details"; the number the whole comparison turns on is in there.
What ships with it
15 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assets/compare/__init__.py 0 B runs code
- assets/compare/__main__.py 3.9 KB runs code
- assets/compare/assemble.py 3.3 KB runs code
- assets/compare/axes.py 6.8 KB runs code
- assets/compare/fulltext.py 4.7 KB runs code
- assets/compare/report.py 3.6 KB runs code
- assets/compare/resolve.py 4.2 KB runs code
- assets/run-compare.py 709 B runs code
- assets/tests/test_assemble.py 3.9 KB runs code
- assets/tests/test_axes.py 6.5 KB runs code
- assets/tests/test_fulltext.py 5.3 KB runs code
- assets/tests/test_report.py 3.4 KB runs code
- assets/tests/test_resolve.py 3.6 KB runs code
- assets/tests/test_stdlib_only.py 889 B runs code
- reference/axes.md 3.6 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 110 lines · 91 tokens per session scan A 304b62cb0f46
comparing-papers is a skill published in the GitHub repository chgagne/claude-skills-research (4 stars, last pushed 6d ago), licensed MIT. It adds 91 tokens to every session and 1,316 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
paper-writer
Medical/scientific paper writing workflow skill. Manages the full pipeline from literature search to submission-ready manuscript. Creates and manages a project directory with IMRAD-format section files, literature matrix, reference management, and quality checklists. Supports both English and Japanese papers.…
food-research
Run a comprehensive, multi-source literature and evidence-synthesis workflow for food & nutrition science. Use when the user wants to research a food/nutrition topic in depth, do a literature review, build an evidence brief, screen and synthesize many sources, verify citations, or scope a systematic review.…
food-paper
Multi-subagent manuscript system for food & nutrition science covering the whole research process: understand the field, frame research questions, curate and analyze data, run statistics, build figures and tables, construct the discussion, draft, polish, and self-review — journal-aware throughout. Includes a…
food-pipeline
Master orchestrator for the whole food & nutrition research-to-publication workflow. Coordinates the specialist skills — each with its own subagent set — into one governed path: journal selection, research (food-research / food-deep-research), writing & analysis (food-paper), figures (food-figure), peer review…
food-deep-research
General-purpose deep research that produces a fully written, source-validated literature review on any question: scope it, design the method, discover and screen sources by journal ranking, validate every source, extract and verify evidence, synthesize, stress-test, then write and format the review (APA 7.0 by…
food-figure
Comprehensive figure system for food & nutrition manuscripts: analyzes the user's data, recommends the best figure(s) to make, then produces submission-grade graphics in Python or R at the target journal's spec. Handles all common scientific figure types (bar/box/violin, line/kinetic, scatter/regression, Bland–Altman…