paper-review

Render LaTeX or Markdown sources into an interactive web review interface.

19|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/angadhn/botference --skill paper-review-angadhn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: paper-review
Source: https://github.com/angadhn/botference/tree/main/.agents/skills/paper-review
Command: npx skills add https://github.com/angadhn/botference --skill paper-review-angadhn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandoc, and includes scripts (resource) and assets (resource) components.

What problem does it solve?

This skill solves the friction of reviewing technical documents by providing a structured, comment-driven interface that keeps discussions anchored to specific document sections while managing bot-assisted edits.

Core Features & Use Cases

  • Anchored Review: Enables Google-Docs-style margin comments on rendered LaTeX or Markdown sources, ensuring feedback is always context-aware.
  • Bot-Human Collaboration: Allows agents like Claude and Codex to participate in review rounds, suggest changes, and resolve tasks via a strict, social routing protocol.
  • Atomic Commit Workflow: Provides a safe, deterministic apply-and-commit mechanism for document changes, ensuring that source edits are always verified and human-approved.

Quick Start

Run the botference review command in your terminal followed by the directory path to initialize the review interface for your project.

Frequently Asked Questions about paper-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add anchored margin comments to a LaTeX document for collaborative review?▼

To add anchored margin comments, render your LaTeX or Markdown sources into an interactive web interface. This creates a Google-Docs-style review environment where feedback cards attach directly to specific document sections.

Can AI agents and humans collaborate on the same document review round?▼

Yes, agents and humans can collaborate on document review rounds. The system manages multi-agent and human feedback loops through a strict social routing protocol, allowing bots like Claude to suggest changes and resolve tasks alongside human reviewers.

Do I need Pandoc and Node.js to run a local Markdown review interface?▼

Yes, you need Pandoc to render Markdown or LaTeX sources and a local Node.js environment. Node.js serves the interactive review interface and handles state synchronization, while Pandoc handles the document conversion.

What is the best way to apply suggested edits from a collaborative review back to Markdown source code?▼

The best way to apply suggested edits is using the atomic commit workflow. This mechanism provides a safe, deterministic apply-and-commit process for document changes, ensuring that source edits are always verified and human-approved before integration.

Are there limitations when using bot-assisted edits for technical documentation?▼

The main limitation is that bot-assisted edits require a local Node.js environment and Pandoc, meaning it is not a cloud-native solution. Additionally, all bot-suggested changes must pass through a human-approved atomic commit workflow before altering the source.