Skip to content
Back to search

academic-writing-agents

Claude Code plugin: multi-agent orchestrator with 10 specialist agents for academic writing review, research, drafting, and polishing

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/andrehuang--academic-writing-agents/badge.svg)](/listings/andrehuang--academic-writing-agents)

HTML

<a href="/listings/andrehuang--academic-writing-agents"><img src="/api/listings/andrehuang--academic-writing-agents/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

# Install from the marketplace:
claude plugin install andrehuang-academic-writing-agents

# Or install directly from GitHub:
claude plugin install --url https://github.com/andrehuang/academic-writing-agents

Or test locally during development:

claude --plugin-dir /path/to/academic-writing-agents

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 15 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "name": "andrehuang-academic-writing-agents",
  3. "owner": {
  4. "name": "Haiwen Huang",
  5. "email": ""
  6. },
  7. "plugins": [
  8. {
  9. "name": "academic-writing-agents",
  10. "description": "12 specialist agents and 30 writing principles for academic writing review, research, drafting, polishing, bibliography auditing, and literature surveys, coordinated by a multi-agent orchestrator.",
  11. "source": "./",
  12. "category": "productivity"
  13. }
  14. ]
  15. }
README.md · README · 172 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Academic Writing Agents
  2. A [Claude Code plugin](https://docs.anthropic.com/en/docs/claude-code/plugins) that brings multi-agent review to academic writing — the same "multiple expert reviewers" model that makes peer review work, built into your writing process.
  3. ## Why
  4. The gap between doing great research and writing about it clearly costs real impact. Findings get buried in unclear prose, reviewers stumble over notation inconsistencies, and figures fail to carry their weight. Most researchers get feedback from one or two people, often late in the process.
  5. This plugin gives you a team of 12 specialist agents — each focused on a different dimension of writing quality — working in parallel, available at any point in the writing process.
  6. ## Design Philosophy
  7. Informed by Michael Black's "Writing a Good Scientific Paper" and real thesis/paper writing experience:
  8. - **Parallel expert review.** Multiple specialized agents review simultaneously, each catching what others miss — like having a writing coach, a technical reviewer, a figure specialist, and a bibliography checker all reading at once.
  9. - **Review-then-act.** Always diagnose before fixing. The writing-reviewer identifies issues; the prose-polisher implements fixes. Never blindly edit.
  10. - **30 codified principles.** Not ad-hoc corrections but a systematic framework organized into 6 categories: Structure & Narrative, Prose & Style, Math & Equations, Figures & Tables, Citations & Bibliography, and Process & Meta.
  11. - **Human-in-the-loop.** Agents propose; humans decide. The plugin presents a prioritized plan and the author chooses what to act on.
  12. - **Goal-Problem-Solution rhythm.** From Black's guide — every section should follow the GPS pattern. The plugin itself is structured this way: identify what you want to improve (goal), diagnose what's wrong (problem), deploy the right agents to fix it (solution).
  13. ## How It Works
  14. 1. **Auto-triggers** when Claude detects academic writing context (`.tex` files, thesis chapters, paper drafts).
  15. 2. The `academic` skill reads project conventions from your `.claude/CLAUDE.md`.
  16. 3. Deploys the right specialist agents in parallel.
  17. 4. Synthesizes findings into a prioritized report (Critical / Important / Minor).
  18. 5. Offers to fix issues via action agents or direct edits.
  19. 6. Iterates with you until you're satisfied.
  20. You can also invoke it manually: `/academic review my introduction`.
  21. ## What's Included
  22. ### Skill
  23. | Skill | Command | Description |
  24. |-------|---------|-------------|
  25. | **Academic** | `/academic <task>` | Orchestrates specialist agents for academic writing tasks |
  26. ### Agents (12 total)
  27. **Review Agents** (read-only analysis):
  28. | Agent | Focus |
  29. |-------|-------|
  30. | `consistency-checker` | Terminology, cross-refs, figure-text-caption alignment, structural coherence |
  31. | `logic-reviewer` | Argument flow, transitions, narrative arc, logical gaps |
  32. | `technical-reviewer` | Math notation, methodology, results validity, citations |
  33. | `writing-reviewer` | Prose clarity, conciseness, grammar, academic tone |
  34. | `latex-layout-auditor` | PDF layout audit — float placement, alignment, sizing |
  35. **Audit Agents** (read + verify):
  36. | Agent | Focus |
  37. |-------|-------|
  38. | `bibliography-auditor` | Bib entry completeness, arXiv updates, title capitalization, venue consistency |
  39. **Research Agents** (read + web):
  40. | Agent | Focus |
  41. |-------|-------|
  42. | `research-analyst` | Related work, novelty assessment, positioning, gap analysis |
  43. | `brainstormer` | Alternative framings, cross-disciplinary connections, research directions |
  44. **Survey Agents** (read + web + write):
  45. | Agent | Focus |
  46. |-------|-------|
  47. | `paper-crawler` | Collects papers from DBLP + OpenAlex APIs, deduplicates, classifies |
  48. **Action Agents** (read + write):
  49. | Agent | Focus |
  50. |-------|-------|
  51. | `prose-polisher` | Rewrites text for clarity, conciseness, flow (edits, not just reports) |
  52. | `section-drafter` | Drafts LaTeX sections, paragraphs, transitions, captions, abstracts |
  53. | `latex-figure-specialist` | Creates/adjusts TikZ/pgfplots figures, manages placement and layout |
  54. ### 30 Principles
  55. Organized into 6 categories:
  56. **A. Structure & Narrative** — A1 Recursive Consistency, A2 Logical Chaining, A3 Definition Order, A4 Paragraph Closers, A5 Claim-First, A6 GPS Rhythm, A7 The Nugget
  57. **B. Prose & Style** — B1 Enumerations, B2 Negation-Contrast, B3 Colloquial Terms, B4 Thesis Voice, B5 One Idea/Sentence, B6 Calibrated Confidence, B7 Ruthless Conciseness, B8 AI-Tell Detection
  58. **C. Math & Equations** — C1 Math for Clarity, C2 Triple Explanation, C3 Equation-Code Correspondence
  59. **D. Figures & Tables** — D1 Active Figures, D2 Cross-Reference Floats, D3 Figure-Text-Caption Consistency, D4 One Message, D5 Interpret Figures, D6 Row Alignment, D7 Caption Self-Sufficiency
  60. **E. Citations & Bibliography** — E1 Cite Named Entities, E2 Citation Completeness, E3 Bibliography Hygiene
  61. **F. Process & Meta** — F1 Strategic Limitations, F2 Negation-Contrast Audit
  62. ## Usage
  63. The plugin **auto-triggers** when Claude detects academic writing context — `.tex` files, thesis chapters, paper drafts, bibliography work. Just describe what you need in natural language:
  64. ```
  65. Review chapter 3 for consistency and technical correctness
  66. Draft a transition paragraph from the method section to experiments
  67. Polish the abstract in parts/abstract.tex
  68. Audit the bibliography for missing fields and arXiv updates
  69. How does our approach compare to recent test-time adaptation papers?
  70. Survey recent work on test-time adaptation from top venues 2023–2025
  71. Plan the structure for a related work section on domain generalization
  72. Prepare this paper for NeurIPS submission — full review pipeline
  73. ```
  74. You can also invoke the skill explicitly with `/academic <task>` if auto-triggering doesn't activate.
  75. ## Installation
  76. ```bash
  77. # Install from the marketplace:
  78. claude plugin install andrehuang-academic-writing-agents
  79. # Or install directly from GitHub:
  80. claude plugin install --url https://github.com/andrehuang/academic-writing-agents
  81. ```
  82. Or test locally during development:
  83. ```bash
  84. claude --plugin-dir /path/to/academic-writing-agents
  85. ```
  86. ## Review Persistence & Coordination (NEW in v2.1)
  87. The orchestrator now **persists review findings** and supports **incremental review**:
  88. - **Review results saved to `.review/`**: After synthesis, aggregated findings (Critical/Important/Minor) are written to `.review/YYYY-MM-DD-<scope>.md` in your project. This means review results survive across sessions.
  89. - **Incremental review**: Before deploying reviewers, the orchestrator checks if unchanged files were already reviewed. If nothing changed since the last review, it presents the prior findings and asks if you want a fresh review or want to proceed to fixes.
  90. - **Cross-agent findings sharing**: When action agents (prose-polisher, section-drafter) are deployed to fix issues, they receive the specific reviewer concerns relevant to their scope. This ensures fixes address flagged issues rather than applying generic improvements.
  91. These features work with [Claude-Claw](https://github.com/andrehuang/claude-claw) for session continuity and task management.
  92. ## Customization
  93. ### Project-level agents
  94. Add domain-specific agents to your project's `.claude/agents/` directory. The plugin auto-discovers them and includes them in deployment plans.
  95. ### Project-level conventions
  96. Add a `.claude/CLAUDE.md` to your project with:
  97. - Document structure (chapters, files, build commands)
  98. - LaTeX conventions (macros, figure paths, citation style)
  99. - Author writing style profile (voice, tone, patterns to match)
  100. All agents read this before starting, so they adapt to your project's specific conventions.
  101. ### Extending the agent roster
  102. Add new agents to `~/.claude/agents/` (global) or `<project>/.claude/agents/` (project-level). Each agent is a Markdown file with YAML frontmatter specifying name, description, tools, and model. The plugin picks them up automatically.
  103. ## Acknowledgments
  104. The principles and workflows in this plugin draw on:
  105. - Michael Black's "Writing a Good Scientific Paper" — the GPS rhythm, the nugget concept, and the emphasis on figures as first-class citizens
  106. - Real thesis supervision experience — patterns observed across multiple rounds of advisor feedback
  107. - Peer review experience — what reviewers actually flag and what they miss
  108. ## License
  109. MIT

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.