Agent skill

Medical Review Writer Architect

by aipoch in aipoch/medical-research-skills

A multi-stage workflow for writing long-form medical reviews; used when the user needs to build an outline based on PubMed literature, write chapter by chapter, perform supplementary searches, and…

MITAuto-check passedResearch & Science

Install Medical Review Writer Architect

skills CLI
$ npx skills add aipoch/medical-research-skills --skill medical-review-writer-architect -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install aipoch/medical-research-skills medical-review-writer-architect --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/aipoch/medical-research-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/'scientific-skills/Academic Writing/medical-review-writer-architect' .claude/skills/medical-review-writer-architect && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
medical-review-writer-architect
GitHub stars
1.9k
Token cost
~2.7k tokens
SKILL.md length
1,264 words
Files
16 (incl. scripts, references, assets)
Skills in repo
578
Repo updated
First seen
Licence
MIT

At a glance

A multi-stage workflow for writing long-form medical reviews; used when the user needs to build an outline based on PubMed literature, write chapter by chapter, perform supplementary searches, and…

  • Works in 6 steps: Outline Construction (Outline) → Data Retrieval and Library Construction → Chapter-by-Chapter Writing → …
  • Needs to build an outline based on PubMed literature
  • SKILL.md covers When to Use, When Not to Use, Required Inputs and Output Contract, plus 7 more sections
  • Runs Python scripts from its folder; calls python

What it does

Medical Review Writer Architect is an agent skill from aipoch/medical-research-skills. A multi-stage workflow for writing long-form medical reviews; used when the user needs to build an outline based on PubMed literature, write chapter by chapter, perform supplementary searches, and format citations; input is the review topic and project directory, output is a c...

Its SKILL.md is about 2.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 18 other files, including scripts, reference files and assets (for example `POLISH_CHANGELOG.md`, `README.md` and `assets/Review_Final.md`).

It sits in Research & Science, covering Performance reviews, Academic paper search and Citation management. It works with PubMed. The repository describes itself as: Hundreds of agent skills for medical research, including protocol design, data analysis, evidence insights, and academic writing. The licence is MIT.

When your agent uses it

  • Needs to build an outline based on PubMed literature
  • Write chapter by chapter
  • Perform supplementary searches
  • Format citations

Example prompts

  • “/medical-review-writer-architect”

Requirements

  • Python 3

Workflow steps

6 steps, taken from the step headings in SKILL.md.

  1. Outline Construction (Outline)
  2. Data Retrieval and Library Construction
  3. Chapter-by-Chapter Writing
  4. Conclusion Writing
  5. Abstract and Finalization
  6. Citation Formatting

What it can do on your machine

Read from SKILL.md and the folder at commit 686e09d. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 8 files in scripts/ (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Medical Review Writer Architect loads about 2.7k tokens when it runs, and up to ~8.4k if it reads all its reference files. Until then it costs about 78 tokens; SKILL.md has 1,264 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~78
When it runs · the whole SKILL.md, loaded when a task matches
~2.7k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~8.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from aipoch/medical-research-skills at commit 686e09d, republished under its MIT licence (© aipoch). 1,264 words, ~2,742 tokens.

Download SKILL.mdSave it as .claude/skills/medical-review-writer-architect/SKILL.md (or your agent's skills folder). This skill also uses 15 other files; get the full folder from GitHub.
name
medical-review-writer-architect
description
A multi-stage workflow for writing long-form medical reviews; used when the user needs to build an outline based on PubMed literature, write chapter by chapter, perform supplementary searches, and format citations; input is the review topic and project directory, output is a c...
license
MIT
author
AIPOCH

Source: https://github.com/aipoch/medical-research-skills

Medical Review Writer (Architect)

When to Use

  • The user explicitly wants to write a medical review, narrative review, or long-form review based on PubMed evidence.
  • The user accepts a multi-stage process of outline, search, chapter-by-chapter writing, verification, and formatting.
  • The user needs real intermediate files, rather than generating the final long document in one shot within the conversation.

When Not to Use

  • When the user only needs a summary, single paragraph review, or brief background introduction, do not launch the full workflow.
  • When the user has no topic, target audience, language requirements, or project directory, do not start writing directly — first collect the minimum inputs.
  • When the current task does not allow creating or modifying intermediate files, do not use this skill, as its core workflow depends on real files and state machine checks.

Required Inputs

FieldRequiredFormat/SourceExampleIf Missing
topicYesTextTP53 in lung cancerStop and ask user to provide
project_dirYesLocal directoryTP53/Confirm working directory first
languageNoTextChinese / EnglishDefault to Chinese if not provided
target_scopeNoTextnarrative review / mini reviewExecute as standard long review if not provided

Output Contract

  • Primary output: Review_Final.md and Review_Final_Formatted.md.
  • Intermediate files: outline.md, references.json, introduction.md, section_*.md, abstract.md, conclusion.md.
  • Citation format: Body text uses only [PMID: xxx] as intermediate citations, with final formatting applied later.
  • If the process is interrupted: Must state which stage was completed, what is missing, and mark as PARTIAL.

User Checkpoints

  • After the outline is completed, the user must confirm before entering the search and writing stages.
  • After each chapter passes verification, report the current chapter status and obtain user confirmation before proceeding to the next chapter.

Core Workflow

Strictly follow the six-step process below. Core mechanism: After each script execution, the terminal prints a [State Machine] prompt — strictly read the current state and follow the "next step suggestion."

Stage 1: Outline Construction (Outline)
  1. Analyze the topic and plan 3-5 major chapters (Body Sections).
  2. Critical mandatory requirement: The outline must be detailed down to the sub-heading level (i.e., ### 1.1, ### 1.2, etc.). Each major chapter must have 2-4 pre-planned sub-headings with distinct logical divisions — writing only to the ## 1. Chapter Name level is never permitted.
  3. Create outline.md. Must include: Title, Introduction, Body Sections (including all sub-headings), Conclusion.
  4. Confirm the outline with the user before proceeding to the next step.

Detailed instructions: See references/workflow-details.md Stage 1

Stage 2: Data Retrieval and Library Construction
  1. Design search strategies for each chapter (must include Introduction)
  2. Execute searches: python scripts/pubmed_search.py "QUERY" --section "introduction|section_1|..."
    • Naming convention: --section must strictly correspond to file names (e.g., use introduction NOT Intro, use section_1 NOT Section1_TP53) so subsequent scripts can auto-match
    • Batch retrieval: The script has batch PMID detail fetching enabled for improved efficiency
    • Rate limiting: Automatically controls approximately 1 request/second for stability and API compliance
    • Default count: --max defaults to 20 articles; adjust as needed
    • Safety check: Automatically handles missing PMID or title fields
    • Important: references.json should be saved in the project directory (e.g., TP53/references.json), not the skill directory
  3. Check counts: Run python scripts/count_references.py
    • Target: Ensure total Unique PMIDs in references.json reach at least 150
    • If insufficient, optimize search terms and perform additional searches
  4. Ensure each chapter has adequate literature support

Detailed instructions: See references/workflow-details.md Stage 2

Stage 3: Chapter-by-Chapter Writing
  1. Read chapter references: python scripts/read_references.py "SectionName"
  2. Create introduction.md (Introduction) and section_1.md, section_2.md, etc. for body sections
  3. Format and word count requirements (refer to assets/Review_Final.md):
    • Never write an entire Section as a single long paragraph! Each major chapter (e.g., ## 1. xxx) must be divided into 2-4 specific sub-headings (e.g., ### 1.1 xxx).
    • Content under each sub-heading must be developed into detailed, coherent academic paragraphs — using simple bullet-point shorthand is not permitted.
  4. Literature citation marking: Every objective academic claim needs literature support, using [PMID: xxx] format.
  5. Write-as-you-search (Dynamic Retrieval): If you find insufficient evidence, immediately perform supplementary searches:
    • python scripts/pubmed_search.py "Specific Query" --section "CurrentSection"
  6. Post-writing verification and supplementation:
    • After completing a chapter, verify: python scripts/check_section.py "section_1.md"
    • If the state machine indicates insufficient citations, perform supplementary searches with the --post-write parameter as suggested.
    • If the state machine indicates word count below the predetermined requirement (i.e., < 7000 characters), expand the content depth of each sub-heading as needed.
    • If the state machine detects dense citations (i.e., a single sentence containing > 2 references), expand and revise that sentence to discuss multiple references separately — stacking more than 2 references at the end of one sentence is prohibited.
    • Mandatory requirement: Once verification fully passes (no count/word-count/density warnings), you must report the current chapter's writing status to the user and ask whether to continue — proceed to the next chapter only after receiving explicit user approval.
  7. Document revision mechanism: If revisions are needed to the already-formatted final document, never directly modify the version with hyperlinks. First run python scripts/revert_citations.py Review_Final_Formatted.md -o draft.md to revert to [PMID: xxx] format. Pay attention to the state machine output regarding reference count — if it indicates insufficient references (e.g., < 30), also use pubmed_search.py with the --post-write parameter to supplement with new literature during revision, then re-run the formatting script.
  8. Format reference: The body section portions of assets/Review_Final.md

Detailed instructions:

Show full SKILL.md (395 more words)Show less
Stage 4: Conclusion Writing
  1. Review all section_*.md files
  2. Create conclusion.md: Summarize core findings, echo the introduction, look ahead to the future (do not include citations)
  3. Format reference: The conclusion section of assets/Review_Final.md

Detailed instructions: See references/workflow-details.md Stage 4

Stage 5: Abstract and Finalization
  1. Create abstract.md:
    • Must include the review title (refer to assets/Review_Final.md format)
    • Include: Background, Content, Findings, Conclusion
    • Do not include citations
    • 5-7 keywords, separated by Chinese semicolons (;)
  2. Merge and finalize: python scripts/merge_review.py abstract.md introduction.md section_1.md ... -o Review_Final.md (Note: do not include outline.md)
  3. Format verification: Compare Review_Final.md against assets/Review_Final.md

Detailed instructions: See references/workflow-details.md Stage 5

Stage 6: Citation Formatting
  1. Run: python scripts/format_citations.py Review_Final.md -o Review_Final_Formatted.md
  2. Verify: Check citation links and reference list

Detailed instructions: See references/citation-guide.md

Core Principles

  • Intermediate files: Must create real files; cannot process only in memory
  • Citation standards: Use only [PMID: xxx] format for easy downstream script parsing
  • Language: Default to Chinese for writing; technical terms may retain English
  • Rigor: Every claim in the introduction and body sections should be supported by literature from references.json
  • Citation scope: Use citations only in the Introduction and Body Sections — Abstract and Conclusion do not use citations
  • Diversity: A single reference should ideally be cited no more than 2 times (3 at most); aim to cite as many different references as possible

Resource Files

  • scripts/: Automation scripts (pubmed_search.py, merge_review.py, format_citations.py, etc.)
  • assets/: Output templates (Review_Final.md)
  • references/: Detailed documentation (workflow-details.md, citation-guide.md)

Failure Handling

  • Missing critical input: Explicitly state which fields, files, or identifiers are missing and pause.
  • Script, template, or resource execution failure: Report the failing step, likely cause, and recovery suggestions — do not silently degrade.
  • Partial completion only: Return the verified portion first, then list remaining blockers and suggested next steps.

Input Validation

This skill accepts requests that match the documented purpose of medical-review-writer-architect and include enough context to complete the workflow safely.

Do not continue the workflow when the request is out of scope, missing a critical input, or would require unsupported assumptions. Instead respond:

medical-review-writer-architect only handles its documented workflow. Please provide the missing required inputs or switch to a more suitable skill.

Quick Validation

  • Check that key scripts, templates, or reference file paths this skill depends on exist.
  • Check that the final output contains the core fields, sections, or files specified for this task.
  • Check that results clearly mark assumptions, limitations, and incomplete items.

© aipoch, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 15 other files (scripts, references, assets) in scientific-skills/Academic Writing/medical-review-writer-architect of aipoch/medical-research-skills.

  • SKILL.md
  • .gitignore
  • POLISH_CHANGELOG.md
  • README.md
  • assets/Review_Final.md
  • eval_report_medical-review-writer-architect_result.json
  • references/citation-guide.md
  • references/workflow-details.md
  • scripts/check_section.py
  • scripts/count_references.py
  • scripts/format_citations.py
  • scripts/internal_search.py
  • scripts/merge_review.py
  • scripts/pubmed_search.py
  • scripts/read_references.py
  • scripts/revert_citations.py

Open the folder on GitHubat commit 686e09d

Compare with similar skills

Medical Review Writer Architect next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Medical Review Writer Architect compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Medical Review Writer Architect this skillaipoch/medical-research-skills1.9k—~2.7kAutomated safety check: PassMIT
Literature Reviewneflibata-feng/MyArxiv-Agent12620 repos~5.9kAutomated safety check: NotesMIT
Citation ManagementK-Dense-AI/claude-scientific-writer2.4k2 repos~3.9kAutomated safety check: NotesMIT
Citation Managementneflibata-feng/MyArxiv-Agent12619 repos~8.1kAutomated safety check: NotesMIT
Academic Search and Citation RouterYuan1z0825/nature-skills47k—~884Automated safety check: PassApache-2.0
Nature Academic Searchjing1312/nature-figure-skill171—~1.3kAutomated safety check: NotesMIT

Similar skills

  • Literature Review

    neflibata-feng/MyArxiv-Agent

    Conduct comprehensive, systematic literature reviews using multiple academic databases (PubMed, arXiv, bioRxiv, Semantic Scholar, etc.).

    126 GitHub starsUsed in 20 repos~5.9k tokens
    Research & ScienceAuto-check: notes
  • Citation Management

    K-Dense-AI/claude-scientific-writer

    Finds papers in OpenAlex, PubMed and Google Scholar, turns DOIs, PMIDs and arXiv IDs into clean BibTeX, and validates citations for a manuscript or thesis.

    2.4k GitHub starsUsed in 2 repos~3.9k tokens
    Research & ScienceAuto-check: notes
  • Citation Management

    neflibata-feng/MyArxiv-Agent

    Comprehensive citation management for academic research. An agent skill from neflibata-feng/MyArxiv-Agent.

    126 GitHub starsUsed in 19 repos~8.1k tokens
    Research & ScienceAuto-check: notes
  • Academic Search and Citation Router

    Yuan1z0825/nature-skills

    Finds papers across literature sources, verifies and converts citations, builds MeSH strategies and audits independent citations of a paper.

    47k GitHub stars~884 tokensUpdated today
    Research & ScienceAuto-check passed
  • Nature Academic Search

    jing1312/nature-figure-skill

    Multi-source literature search, citation verification, MeSH search strategy, citation file management (.nbib/.ris/.bib conversion), and reference management (BibTeX, related articles, ID conversion)…

    171 GitHub stars~1.3k tokensUpdated 1 mo ago
    Research & ScienceAuto-check: notes
  • Reference Checker

    Liuxiangjian-ai/reference-checker-skill

    Exhaustively verify English and Chinese manuscript references before journal submission.

    140 GitHub stars~5.4k tokensUpdated 4 mo ago
    Research & ScienceAuto-check passed

More from aipoch/medical-research-skills

All 578 skills in this repo
  • Academic Poster Generator

    aipoch/medical-research-skills

    Complete workflow for generating academic research posters from PDF literature; use when you need to extract paper content from PDFs and produce a LaTeX-based poster…

    1.9k GitHub stars~2.2k tokensUpdated 24 days ago
    Auto-check passed
  • Diagnostic Study Quality Assessment Quadas

    aipoch/medical-research-skills

    Analyzes clinical diagnostic accuracy studies for bias using the QUADAS-2 tool.

    1.9k GitHub stars~1.4k tokensUpdated 24 days ago
    Auto-check passed
  • Exploratory Data Analysis

    aipoch/medical-research-skills

    Perform comprehensive exploratory data analysis on scientific data files across 200+ file formats.

    1.9k GitHub stars~3.7k tokensUpdated 24 days ago
    Auto-check passed
  • Iso Certification

    aipoch/medical-research-skills

    A toolkit for preparing ISO 13485:2016 certification documentation for medical device QMS.

    1.9k GitHub stars~1.8k tokensUpdated 24 days ago
    Auto-check passed
  • Journal Skills

    aipoch/medical-research-skills

    Recommends target journals for manuscript submission by analyzing the paper topic/abstract and the journal distribution of similar PubMed literature; use when users ask for journal…

    1.9k GitHub stars~1.7k tokensUpdated 24 days ago
    Auto-check passed
  • Latex Posters

    aipoch/medical-research-skills

    Creates academic-poster writing packages for LaTeX using beamerposter, tikzposter, or baposter.

    1.9k GitHub stars~1.3k tokensUpdated 24 days ago
    Auto-check passed

Works with

Questions about Medical Review Writer Architect

What does Medical Review Writer Architect do?

A multi-stage workflow for writing long-form medical reviews; used when the user needs to build an outline based on PubMed literature, write chapter by chapter, perform supplementary searches, and…. Medical Review Writer Architect is an agent skill from aipoch/medical-research-skills. A multi-stage workflow for writing long-form medical reviews; used when the user needs to build an outline based on PubMed literature, write chapter by chapter, perform supplementary searches, and format citations; input is the review topic and project directory, output is a c...

When should I use Medical Review Writer Architect?

Medical Review Writer Architect fits situations like: needs to build an outline based on PubMed literature; write chapter by chapter; perform supplementary searches; format citations.

How do I install Medical Review Writer Architect in Claude Code?

Run `npx skills add aipoch/medical-research-skills --skill medical-review-writer-architect -a claude-code`. Or copy the skill folder (scientific-skills/Academic Writing/medical-review-writer-architect in aipoch/medical-research-skills) into .claude/skills/medical-review-writer-architect in your project. Claude Code loads it when a task matches its description.

How do I install Medical Review Writer Architect in Codex?

Run `npx skills add aipoch/medical-research-skills --skill medical-review-writer-architect -a codex`. Or copy the skill folder (scientific-skills/Academic Writing/medical-review-writer-architect in aipoch/medical-research-skills) into .agents/skills/medical-review-writer-architect in your project. Codex loads it when a task matches its description.

Can I use Medical Review Writer Architect in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add aipoch/medical-research-skills --skill medical-review-writer-architect -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/medical-review-writer-architect, .gemini/skills/medical-review-writer-architect, .github/skills/medical-review-writer-architect and .opencode/skills/medical-review-writer-architect in your project.

What does Medical Review Writer Architect need to run?

Going by SKILL.md and its folder, Medical Review Writer Architect needs Python for the scripts in its folder and the command-line tools its instructions call (python). Our summary lists: Python 3.

Does Medical Review Writer Architect access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Medical Review Writer Architect safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Medical Review Writer Architect use?

Medical Review Writer Architect is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Medical Review Writer Architect use?

About 2.7k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 5.6k tokens, read only when the agent opens those files.

What are the alternatives to Medical Review Writer Architect?

Skills that share tags, products or a category with Medical Review Writer Architect: Literature Review (neflibata-feng/MyArxiv-Agent, 126 stars), Citation Management (K-Dense-AI/claude-scientific-writer, 2.4k stars), Citation Management (neflibata-feng/MyArxiv-Agent, 126 stars) and Academic Search and Citation Router (Yuan1z0825/nature-skills, 47k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Medical Review Writer Architect?

aipoch (a GitHub organization) maintains it in aipoch/medical-research-skills, which has 1,937 GitHub stars. The repository holds 578 skills in this directory. The repository was last updated on September 17, 2026.

Source: aipoch/medical-research-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.