# grant-proposal-assistant

Guides writing and review of NIH R01/R21/K, NSF and foundation grant proposals by reading them the way a reviewer would.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 350 installs · ★ 186
- Author: lyndonkl
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/lyndonkl/claude/tree/HEAD/skills/grant-proposal-assistant/SKILL.md
- Updated: 2026-09-01
- HTML: https://affpapa.org/en/nevibecode/skills/grant-proposal-assistant

## What it does

The skill walks a proposal through eight steps: pin down the mechanism and its constraints, audit the four core questions, review the one-page Specific Aims, then the Significance, Innovation and Approach sections, check alignment with a reviewer's view, and verify compliance. Each section has its own framework. The Aims page follows a fixed shape: hook, gap, hypothesis, aims, impact. In Approach, every aim gets a rationale, a strategy, expected outcomes, pitfalls and alternatives.

The guiding idea is to read the proposal as a non-expert reviewer would. So the skill checks whether significance lands without deep domain knowledge and whether impact statements are prominent. It also flags weak spots: an unclear hypothesis, a missing gap statement, vague claims about what changes if the work succeeds.

Supporting files live in `resources/methodology.md`, `resources/template.md` and the scoring rubric `resources/evaluators/rubric_grant_proposal.json`. The rubric sets a minimum average score of 3.5. This is not the skill for manuscripts, fellowship personal statements or recommendation letters — those have their own entries in the set. Check page limits early, because they differ by mechanism: R01 runs 12 pages, R21 only 6, and K is 12 pages plus a career development component.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add lyndonkl/claude --skill grant-proposal-assistant -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/grant-proposal-assistant/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/grant-proposal-assistant/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add lyndonkl/claude --skill grant-proposal-assistant -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/grant-proposal-assistant/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add lyndonkl/claude --skill grant-proposal-assistant -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/grant-proposal-assistant/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install lyndonkl/claude grant-proposal-assistant
```

2. **Manually: put the files in place**
- File: `.github/skills/grant-proposal-assistant/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add lyndonkl/claude --skill grant-proposal-assistant -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/lyndonkl/claude --path skills/grant-proposal-assistant
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/grant-proposal-assistant/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add lyndonkl/claude --skill grant-proposal-assistant -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/lyndonkl/claude/tree/HEAD/skills/grant-proposal-assistant/SKILL.md

## When to use it
- turn a vague study plan into a testable, falsifiable hypothesis
- state the knowledge gap and why it matters now
- draft the Specific Aims page from hook through impact
- replace generic claims in the Innovation section with specifics
- break each aim into rationale, strategy, pitfalls and alternatives

## Example prompts
- Audit my R01 proposal against the four core questions and flag where the hypothesis is vague
- Review my Specific Aims page: is the hook, gap and impact clear enough?
- Read the Approach section as a reviewer and point out where alternatives are missing

## Similar
- [research-grants](https://affpapa.org/en/nevibecode/skills/k-dense-ai-research-grants.md) — skill · 2000 installs
- [k2f-grant-proposal](https://affpapa.org/en/nevibecode/skills/k2f-grant-proposal.md) — skill · 2400 installs
- [grant-proposal](https://affpapa.org/en/nevibecode/skills/wanshuiyin-grant-proposal.md) — skill · 560 installs
- [grants](https://affpapa.org/en/nevibecode/skills/grants.md) — skill · 355 installs
- [academic-humanizer](https://affpapa.org/en/nevibecode/skills/academic-humanizer.md) — skill · 539 installs
- [award-application](https://affpapa.org/en/nevibecode/skills/award-application.md) — skill · 954 installs

From the same set: [lyndonkl/claude](https://affpapa.org/en/nevibecode/by/lyndonkl/claude.md) — 5 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md