grants

Generates NIH funding strategy reports as Word documents using Consensus and RePORTER data.

25.3k|3.6k|Updated Oct 19, 2025
One-click install
npx skills add https://github.com/alirezarezvani/claude-skills --skill grants
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grants
Source: https://github.com/alirezarezvani/claude-skills/tree/main/research/grants/skills/grants
Command: npx skills add https://github.com/alirezarezvani/claude-skills --skill grants

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires docx, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Clinical researchers struggle to identify the right NIH institutes, mechanisms, and study sections for their research ideas, often wasting months on misaligned applications. This Skill turns a research idea into a strategic NIH funding overview as an editable .docx, grounded in live Consensus literature searches and RePORTER project data.

Core Features & Use Cases

  • Structured Intake: Six forcing questions lock down research idea, career stage, preliminary data, environment, submission posture, and institute targets before any search runs.
  • Positioning Analysis: Runs five sequential Consensus searches (established, stakes, current approaches, adjacent methods, gaps) and drafts Significance/Innovation language with quotable gap citations.
  • Institute & Mechanism Mapping: Queries NIH RePORTER via POST to tally funding institutes and study sections, discovers NOSIs, and matches career stage × scope × preliminary data to mechanisms (F31 through U01) via a rule-based matcher.
  • Audited DOCX Output: Produces a 9-section Word document with hyperlinked references, submission timelines, a mandatory program officer recommendation, and a full audit log of queries sent, results received, and results cited.
  • Use Case: A postdoc with pilot data asks "find grants for my sepsis prediction research" and receives a .docx naming NHLBI as the top institute, recommending K99/R00, listing active NOSIs, and providing a program officer contact plan.

Quick Start

Ask the agent to find NIH grant opportunities for your research idea and answer the six intake questions about your career stage and preliminary data.

Frequently Asked Questions about grants

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

FAQPage Schema
How do I find NIH grants for my research idea?

Describe your research idea and answer six intake questions covering career stage, preliminary data, environment, and submission posture. The skill then searches Consensus and NIH RePORTER, maps institutes and study sections, and delivers a strategic .docx with mechanism recommendations.

What NIH funding mechanism matches my career stage?

Mechanism matching uses career stage, project scope, and preliminary data together, not career stage alone. The mechanism_matcher.py script maps combinations to options like F31, K99/R00, R21, R01, or U01 with budget and rationale for each.

Does this skill cover non-NIH funders like PCORI or DOD?

No, the scope is NIH-only. Non-NIH funders such as PCORI, DOD CDMRP, VA, and private foundations are flagged at intake as out of scope and are not searched or recommended.

Why can't web_fetch be used to query NIH RePORTER?

The RePORTER API v2 is POST-only, while web_fetch performs GET requests, so queries would silently fail. The skill uses bash curl POST requests with JSON criteria including fiscal years and advanced text search operators.

What happens when Consensus or RePORTER queries fail?

Failures trigger a 3-second wait and one retry, then are logged in the audit section. After three consecutive failures across tools the run stops and alerts the researcher; missing data is never filled from training knowledge.