loading…
Search for a command to run...
loading…
Medical Writer's AI Toolkit — 33 expert prompts for pharma medical writing as MCP tools
Medical Writer's AI Toolkit — 33 expert prompts for pharma medical writing as MCP tools
The Medical Writer's AI Toolkit — 34 expert prompts as callable MCP tools. Built by a CMPP-certified medical writer with a PhD and 10+ years in pharma. Gives Claude and any MCP-compatible AI agent direct access to publication-ready prompt frameworks for every stage of the pharma medical writing workflow.
| Tier | Tools | Access |
|---|---|---|
| Free | 8 starter tools | $0/mo · 25 requests · install and use immediately |
| Pro | All 34 tools + ADAPT framework | $39/mo ($374/yr) · 1,000 requests |
| Business | All 34 tools + ADAPT framework | $89/mo ($854/yr) · 5,000 requests |
| Ultra | All 34 tools + ADAPT framework | $129/mo ($1,238/yr) · 10,000 requests |
Free tier includes 8 starter tools (manuscript outline, abstracts, and core IMRaD section drafting). Paid tiers unlock the full Manuscript Toolkit — all 26 advanced tools plus the ADAPT framework.
| Tool | Description |
|---|---|
get_manuscript_outline_prompt |
Create a detailed manuscript outline with section word counts |
get_structured_abstract_prompt |
Draft a CONSORT/STROBE-aligned structured journal abstract |
get_unstructured_abstract_prompt |
Draft a narrative (unstructured) journal abstract |
get_introduction_section_prompt |
Draft a manuscript Introduction section |
get_methods_section_prompt |
Draft a manuscript Methods section |
get_results_section_prompt |
Draft a manuscript Results section |
get_discussion_section_prompt |
Draft a manuscript Discussion section |
list_all_tools |
List every tool available in the toolkit |
Chapter 1 — Abstract writing
Chapter 2 — Manuscript development
Chapter 3 — Peer review
Chapter 4 — Congress abstracts and posters
Chapter 5 — Slide deck narratives
Chapter 6 — Publication planning
Chapter 7 — Author collaboration
Chapter 8 — Editing, QC and style
Chapter 9 — Career and professional development
Bonus — ADAPT framework
git clone https://github.com/pubspro/medwriter-mcp.git
cd medwriter-mcp
pip install mcp
Edit %APPDATA%\Claude\claude_desktop_config.json (Windows) or ~/Library/Application Support/Claude/claude_desktop_config.json (macOS):
{
"mcpServers": {
"medwriter-mcp": {
"command": "python",
"args": ["/absolute/path/to/medwriter-mcp/server.py"]
}
}
}
Restart Claude Desktop. All tools appear automatically.
claude mcp add medwriter-mcp python /absolute/path/to/medwriter-mcp/server.py
Once connected, ask Claude:
Use get_structured_abstract_prompt to draft an abstract for a Phase 3 RCT
of pembrolizumab in NSCLC. Primary endpoint: PFS HR 0.58 (95% CI 0.46–0.72, p<0.001).
Use get_gap_analysis_prompt to identify publication gaps for a BTK inhibitor
in CLL based on the current literature landscape.
Use get_congress_abstract_prompt to draft an ASH abstract for a study
of venetoclax plus obinutuzumab in previously untreated CLL.
Use get_promotional_language_check_prompt to review this slide deck excerpt
before MLR submission.
Use apply_adapt_framework to rewrite this abstract for a nursing audience
at a health literacy reading level.
Every prompt is structured so you never need to input unpublished, confidential, or proprietary data. Tools marked 🔒 carry elevated data-sensitivity risk and include inline guidance.
Never input into any public AI tool:
Always comply with ICMJE, GPP4, and your organization's AI use policies.
Pair with other pubspro MCP servers for a complete pharma AI workflow:
pip install mcp)MIT
Run in your terminal:
claude mcp add medwriter-mcp -- npx Yes, Medwriter MCP is free — one-click install via Unyly at no cost.
No, Medwriter runs without API keys or environment variables.
Self-hosted: the server runs locally on your machine via the install command above.
Open Medwriter on unyly.org, pick your client tab (Claude Desktop, Claude Code, Cursor) and press Install — the config is generated automatically, no JSON editing.
CSA PROJECT - FZCO © 2026 IFZA Business Park, DDP, Premises Number 31174 - 001
Security
Low riskAutomated heuristic from public metadata — not a security guarantee.