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 Ingar30/reviewer --skill review-papergit clone --depth 1 https://github.com/Ingar30/reviewerWrote 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/ingar30/reviewer/review-paper)<a href="https://agentmods.dev/skills/ingar30/reviewer/review-paper"><img src="https://agentmods.dev/badge/skills/ingar30/reviewer/review-paper/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/ingar30/reviewer/review-paper"><img src="https://agentmods.dev/badge/skills/ingar30/reviewer/review-paper.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 | $0.00049 | $0.01427 |
| Opus 5 | $0.00024 | $0.00714 |
| Sonnet 5 | $0.00010 | $0.00285 |
| Haiku 4.5 | $0.00005 | $0.00143 |
Grade A, and why
review-paper 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 today.
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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review this paper
Deliver the complete report as a downloadable PDF, with a brief plain-language
summary. Preserve the editor's report.md as the authoritative source and an optional
download; the PDF is a deterministic presentation, not a rewritten review.
Use native ChatGPT Work subagents for parser preflight, the applicability
router, each selected reviewer, and the editor. This skill explicitly requests that
delegation. The parent orchestrates and validates; only the editor authors the report.
Start with the user experience
For a PDF-only export of an existing completed Reviewer report, go directly to PDF-first delivery. Do not start new audits, spend agent allowance, or resume an incompatible older checkpoint just to change the file format. Clearly distinguish exporting a supplied report from independently validating its scientific conclusions.
- For a resume request, resolve the existing review folder or attached checkpoint first; a valid checkpoint already contains the source PDF, so do not request it again. For a new review, resolve the attached PDF from the host's actual attachment/file tools; do not guess an upload path. If none is attached, ask for it. If several are attached and the intended paper is ambiguous, ask which one; do not combine unrelated papers.
- Default to Full. Honor an explicit Lite or resume request without asking the user to configure paths or reviewers. Briefly announce the mode and warning: "This uses your shared ChatGPT Work/Codex allowance, with no API key. A Full review can use substantial allowance and may pause at your limit. I will save completed audits so you can resume." For Lite add: "Lite keeps the same audit coverage with lower requested reasoning effort; its quality and savings are unbenchmarked." Proceed with the requested mode; do not require a redundant confirmation.
- Read the bundled canonical Paper Reviewer guidance in full for the reviewing methodology, evidence rules and report requirements. Then read the host orchestration procedure. Check native subagents, writable files, Python dependencies, PDF access, and image viewing before substantive review. Discover web search availability.
- Use the packaged
runtime/scripts/work_plugin.pyto prepare and checkpoint the run. All runtime resources are bundled beside this skill; never assume access to the source Git repository or a particular operating system. Keep the installed plugin read-only and use a fresh host-writable folder for each paper review. If the user requests cloud-only execution, establish the host from the actual task/host context before processing. Do not borrow a local checkout, interpreter, or remote connection to the user's computer. An unknown host is not a verified cloud host; explain the limitation instead of silently switching to local work.
What ships with it
44 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.
- agents/openai.yaml 219 B
- references/orchestration.md 14 KB
- runtime/bundle_manifest.json 4.5 KB
- runtime/config/reviewers.json 6.0 KB
- runtime/LICENSE.md 1.0 KB
- runtime/prompts/templates/abstract_conclusion_consistency_audit.txt 1.5 KB
- runtime/prompts/templates/claim_evidence_audit.txt 2.0 KB
- runtime/prompts/templates/crossref_audit.txt 1.2 KB
- runtime/prompts/templates/data_availability_replication_audit.txt 2.8 KB
- runtime/prompts/templates/design_randomization_audit.txt 2.2 KB
- runtime/prompts/templates/economic_magnitude_audit.txt 1.9 KB
- runtime/prompts/templates/editor_report.txt 13 KB
- runtime/prompts/templates/grammar_audit.txt 1.8 KB
- runtime/prompts/templates/identification_audit.txt 1.7 KB
- runtime/prompts/templates/institutional_context_audit.txt 2.4 KB
- runtime/prompts/templates/limitations_external_validity_audit.txt 1.2 KB
- runtime/prompts/templates/literature_audit.txt 1.8 KB
- runtime/prompts/templates/model_equation_audit.txt 1.8 KB
- runtime/prompts/templates/numerical_audit.txt 2.5 KB
- runtime/prompts/templates/parser_quality_audit.txt 5.7 KB
- runtime/prompts/templates/power_multiple_testing_audit.txt 2.1 KB
- runtime/prompts/templates/reference_audit.txt 1.6 KB
- runtime/prompts/templates/reviewer_contract.txt 2.5 KB
- runtime/prompts/templates/reviewer_selection.txt 5.1 KB
- runtime/prompts/templates/robustness_audit.txt 2.1 KB
- runtime/prompts/templates/sample_construction_audit.txt 2.4 KB
- runtime/prompts/templates/source_consistency_audit.txt 3.7 KB
- runtime/prompts/templates/theory_logic_audit.txt 2.2 KB
- runtime/requirements.txt 87 B
- runtime/reviewer-guidance/paper-reviewer.md 7.5 KB
- runtime/schemas/reviewer_output.schema.json 5.1 KB
- runtime/schemas/reviewer_selection.schema.json 1.5 KB
- runtime/scripts/build_editor_input.py 25 KB runs code
- runtime/scripts/check_final_report.py 8.5 KB runs code
- runtime/scripts/normalize_review_outputs.py 24 KB runs code
- runtime/scripts/pipeline_paths.py 1.7 KB runs code
- runtime/scripts/prepare_work_environment.py 5.7 KB runs code
- runtime/scripts/preprocess_pdf.py 147 KB runs code
- runtime/scripts/render_prompts.py 4.2 KB runs code
- runtime/scripts/render_report_pdf.py 11 KB runs code
- runtime/scripts/reviewer_config.py 5.8 KB runs code
- runtime/scripts/reviewer_routing.py 8.1 KB runs code
- runtime/scripts/validate_review_json.py 10 KB runs code
- runtime/scripts/work_plugin.py 49 KB runs code
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.
- today First seen · 102 lines · 49 tokens per session scan A d04c72686877
review-paper is a skill published in the GitHub repository Ingar30/reviewer (170 stars, last pushed yesterday), licensed MIT. It adds 49 tokens to every session and 1,427 once invoked, about $0.0002 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 2026-09-10.
Other skills, from other repositories
compile-latex
Compile a Beamer LaTeX slide deck with XeLaTeX (3 passes + bibtex). Use when user says "compile", "build the slides", "rebuild the PDF", "run latex", "render the tex", or asks why a .tex file isn't producing a PDF. Operates on Slides/.tex.
overleaf
Sync and manage Overleaf LaTeX projects from the command line. Pull projects locally, push changes back, compile PDFs, and download compile outputs like .bbl files for arXiv submissions. Use when working with LaTeX, Overleaf, academic papers, or arXiv.
alterlab-treatment-plans
Generates concise (3-4 page), focused medical treatment plans in LaTeX/PDF format across all clinical specialties — general medical treatment, rehabilitation therapy, mental health care, chronic disease management, perioperative care, and pain management — using SMART goal frameworks, evidence-based interventions with…
alterlab-open-notebook
Run Open Notebook, a self-hosted open-source alternative to Google NotebookLM with a full REST API, for AI-powered research and document analysis. Use when organizing research materials into notebooks, ingesting diverse content sources (PDFs, videos, audio, web pages, Office documents), generating AI-powered notes and…
latex-document
Universal LaTeX document skill: create, compile, and convert any document to professional PDF with PNG previews. Supports resumes, reports, cover letters, invoices, academic papers, theses/dissertations, academic CVs, presentations (Beamer), scientific posters, formal letters, exams/quizzes, books, cheat sheets…
compiledeck
Create and compile beautiful Beamer presentations following the Rhetoric of Decks philosophy. Use when making slides, creating decks, or compiling .tex presentation files.