Grant Writing

Grant Writing Skill

Provides procedural knowledge for drafting NIH and NSF grant proposals with mechanism-specific formatting, section structure, and scientific writing best practices.

Supported Mechanisms

Core Workflow

Step 0: Parse the NOFO/FOA

Before writing anything, read the Notice of Funding Opportunity (NOFO) or Funding Opportunity Announcement (FOA) thoroughly. Extract and record:

• Mechanism and agency: Determines page limits, required sections, and review criteria
• Specific objectives: Many NOFOs list required research areas, populations, or methods
• Page limits: Verify per-section limits (they vary by mechanism and can change between cycles)
• Review criteria and weights: Identify how the review panel will score the application
• Special requirements: Data sharing plans, milestones, letters of support, human subjects provisions
• Eligible PIs: Career stage, citizenship, institutional requirements
• Budget constraints: Direct cost caps, duration limits, cost-sharing restrictions
• Submission deadline and receipt date: Plan backward from the deadline

Store the NOFO URL and key extracted details in NOFO.md within the proposal directory. Flag any requirements that deviate from the standard mechanism template.

Step 1: Plan the page budget

Before drafting, allocate pages across sections. Staying within limits while giving each section adequate space is a critical constraint. Use these as starting points and adjust based on the specific aims and preliminary data available:

R01 (12 pages total strategy):

• Significance: ~2 pages
• Innovation: ~1.5 pages
• Approach: ~8.5 pages (including ~1.5-2p preliminary data, ~5-6p aims, ~1p rigor, ~0.5p timeline)

R21 (6 pages total strategy):

• Significance: ~1 page
• Innovation: ~0.75 pages
• Approach: ~4.25 pages

R03 (6 pages total strategy):

• Significance: ~1 page
• Innovation: ~0.5 pages
• Approach: ~4.5 pages (focus on methods; preliminary data less critical)

K99/R00 (12 pages strategy + 6p candidate + 6p mentoring):

• Research strategy follows R01 proportions
• Candidate section: training background, career trajectory, independence plan
• Mentoring plan: structured activities, timeline, evaluation

F31/F32 (6 pages strategy):

• Follows R21 proportions for strategy
• Applicant background and goals: training rationale, career objectives
• Sponsor and institutional environment: mentor qualifications, training plan, resources

Step 2: Draft the Specific Aims page (NIH) or Project Summary (NSF)

Branch first: is this a small-business application?

If the mechanism is SBIR or STTR (R41, R42, R43, R44), stop and use examples/sbir-specific-aims-template.md instead of the hypothesis-driven template below. Small-business aims are milestone-driven: each aim carries a month window and closes with a quantitative acceptance threshold, with go/no-go conditions where one aim gates the next. Using the hypothesis-driven structure on a small-business application loses points on Approach. See references/sbir-sttr-requirements.md.

NIH Specific Aims (1 page, <650 words):

1. Opening (2-3 sentences): State the problem and critical gap concisely
2. Bold overarching goal: One sentence in bold stating the ultimate objective, followed by innovation/methodology explanation
3. Brief scope statement: Recruitment/context (can merge with goal if space is tight)
4. Aims (2-3): Each titled in bold with an action verb
   • Sub-hypotheses: Italic labels (Hypothesis 1A:), one sentence each
   • After each hypothesis: "We will..." with bold key methodological innovations
5. Expected Impact: Bold header, numbered list of concrete deliverables

See examples/specific-aims-template.md for an annotated template.

NSF Project Summary (1 page):

• Overview paragraph
• Intellectual Merit paragraph
• Broader Impacts paragraph

Step 3: Draft the Research Strategy (NIH) or Project Description (NSF)

NIH Research Strategy (Significance -> Innovation -> Approach):

• Significance (~2p for R01): Problem -> Gap -> Why it matters -> What success enables
• Innovation (~1.5p for R01): What is new -> Why current approaches fail -> Field advancement. Distinguish conceptual, technical, and methodological innovations
• Approach (~8.5p for R01): Preliminary data -> Aim-by-aim breakdown (Rationale -> Methods -> Hypothesis -> Analyses -> Expected outcomes -> Problems/Alternatives) -> Rigor & Reproducibility -> Timeline

Consult references/research-strategy-guidelines.md for detailed section guidance.

NSF Project Description (15 pages):

• Results from Prior NSF Support (within the 15 pages)
• Research Plan
• Broader Impacts activities

NSF CAREER: Research and education plans must be bidirectionally integrated. See references/career-award-guide.md.

Step 4: Draft supporting sections

After the core strategy is drafted, prepare the remaining required documents:

• Facilities & Other Resources: Describe lab space, equipment, computational infrastructure, core facilities, and collaborating sites. Emphasize resources that directly enable the proposed work.
• Equipment: List major equipment available for the project. Distinguish from equipment requested in the budget.
• Data Management & Sharing Plan (DMS, typically 2 pages): Describe what data will be generated, how it will be stored, standards for formatting, timeline for sharing, and any access restrictions. Follow the 2023 NIH DMS Policy or NSF DMP requirements.
• Budget Justification: Justify every budget category. See references/budget-justification-guide.md and examples/budget-justification-template.md.
• Letters of Support: Request from collaborators, consultants, and core facility directors. Each letter should confirm the specific contribution described in the proposal.

For fellowship and career development mechanisms (F31, F32, K08, K23, K99):

• Candidate/Applicant Section: Training history, skills acquired, career goals, and how this award fills gaps
• Mentoring/Training Plan: Structured mentoring activities, timeline, and evaluation criteria
• Sponsor Statement: Mentor's qualifications, commitment, and training record

Step 5: Apply writing style

Consult references/writing-style-guide.md and references/tone-guide.md for established voice conventions. Key principles:

• Direct, active voice with a named actor and a named noun ("We rebuilt Aim 3", not "Aim 3 is rebuilt"; see the habits below)
• Strategic bolding: overarching goal, aim titles, one key innovation per aim, memorable phrases
• Italic hypothesis labels
• No em-dashes (PI style preference, not a universal standard); use commas, semicolons, or parentheses
• Define abbreviations on first use (once per document)
• Quantify when possible (N=24, 6-month, etc.)
• Bold claims backed by technical precision
• First person: "I" for the applicant's vision ("I propose") and own acts on single-investigator mechanisms, "we" for team work

Habits to avoid while drafting. These habits make research prose read as machine-written, and reviewers notice them. Avoid them while drafting; the humanizer pass (patterns 30 to 45) then catches what slips through. The quoted phrases are examples of each habit, not the full list; a sentence that does the same job in other words breaks the same rule.

• Show, do not announce. State the limitation or the result and let it stand. Do not certify your own honesty or rigor ("and we say so", "we say it first", "to be transparent", "we are careful to").
• Say it once. Each sentence adds a fact. Drop tails that restate the purpose (", so a gain must come from what the tags say"), closing sentences that only repeat the paragraph (a sentence that links to an aim or states the gap stays), and a second gloss of a defined term (define it again only in each separately read document). Lead with a purpose ("To separate X from Y, ...") only where a reviewer would ask why.
• Name who acts. Write "We registered the test before seeing any fMRI value", not "The test was registered" or "Hypothesis 1B tests ..."; results predict, differ, or agree, and do not "say", "tell", or "decide". An aim can still be the topic of a sentence ("Aim 2 has two hypotheses"). Use "I" for the applicant's own acts on single-investigator mechanisms.
• Name the noun. When the referent is a sentence back or two nouns could fit, repeat the noun ("CLIP's prediction", not "its prediction"); "this" needs a noun that names something in the sentence just before ("this test"). A pronoun is fine when its referent is the subject of the same clause.
• Literal claims, not slogans. No mirrored contrasts ("fMRI shows where, EEG shows when"), contest metaphors ("beat the null"), colon reveals ("The answer is simple: ..."), or verbs of standing ("rests on", "reads as").
• Answer an objection where it arises. Do not concede and rebut in every paragraph ("X is not the only limit, but ...").
• Plain time and scope. Use "now" or "currently" only to contrast with a stated earlier version or where literal ("currently recruiting"), and state a real constraint ("only", "never") once.
• One idea per sentence. Split chains of "and"; list results strongest first, each with its label.

After drafting, invoke the fully qualified manuscript:humanizer skill through the host's skill-invocation mechanism as a final natural-writing pass. Pay particular attention to:

• Pattern 1 (significance inflation): strip filler ("pivotal moment", "evolving landscape") but keep evidence-backed importance ("affects 1.2M patients annually").
• Pattern 4 (promotional language): "groundbreaking", "transformative", "paradigm-shifting" read as red flags in Innovation sections. Let the methods carry the claim.
• Pattern 24 (excessive hedging): tighten stacked hedges, but keep appropriate uncertainty in Approach risks/alternatives.
• Patterns 17 and 19 (title case, curly quotes): for NIH/NSF, defer to mechanism formatting in Step 6.
• Patterns 30 to 43 (sentence-shape tells such as ", so a gain must come from ...", talking data, and colon reveals): run the skill's quick scan and keep each pattern under one or two hits per 500 words.
• Patterns 44 and 45 (hidden actors and unanchored pronouns): the quick scan counts passive verbs, aims or hypotheses as actors, sentence-initial pronouns, and "it"/"its"; review the hits outside Methods-style text and rewrite with "we" and the named noun.

Step 6: Format for submission

NIH LaTeX formatting:

• Arial 11pt body, 12pt title. Note: Arial is not available in standard LaTeX distributions; use the helvet package for Helvetica, which is an accepted substitute.
• 0.5in margins all sides
• No indentation; 4pt space between paragraphs
• Justified text
• Fit within page limits by trimming content, not reducing spacing

NSF formatting:

• Arial/Helvetica 10pt+, Times New Roman 11pt+, or Palatino 10pt+
• 1 inch margins minimum
• Single-spaced

Step 7: Pre-submission checklist

Before submitting, verify every item:

Content completeness:

• All required sections present per the NOFO
• Page limits respected for every section
• All aims referenced in both the Specific Aims page and the Approach
• Rigor and reproducibility addressed (biological variables, authentication, blinding)
• Timeline with milestones included
• All abbreviations defined on first use
• All figures referenced in text and legible at printed size
• Citations complete and formatted per agency requirements

Budget and administrative:

• Budget matches the narrative (effort levels, equipment, travel consistent with text)
• Budget justification covers every line item
• Biosketches current and formatted correctly (SciENcv for NIH)
• Letters of support included for all named collaborators and consultants
• Data Management and Sharing Plan included
• Facilities and Equipment documents included
• Human subjects or vertebrate animals sections complete (if applicable)

Formatting:

• Font, margins, and spacing meet agency requirements
• No text extends beyond page limits
• PDF renders correctly (no font substitution, no clipped figures)
• File naming follows agency conventions

For resubmissions (A1):

• Introduction page present and within 1-page limit
• All reviewer critiques addressed
• Changes marked in the body of the application
• See references/resubmission-guide.md for detailed guidance

Special Cases

DP2 (New Innovator Award)

• NO specific aims page; write a 10-page essay instead
• Emphasize PI creativity and innovation
• Written for broad audience
• Use catchy component names
• Include "What If" contingency section

Fellowship Mechanisms (F31/F32)

• Applicant background and goals section replaces the standard biosketch narrative
• Sponsor and co-sponsor statements are critical; coordinate closely with mentors
• Training plan must describe specific skills to be acquired
• Institutional environment section should demonstrate access to resources and training opportunities
• See references/nih-requirements.md for page limits and section details

Career Development (K08/K23)

• Candidate section describes prior training and career goals
• Career development plan outlines specific training activities and timeline
• Mentoring plan must name mentors and describe their expertise and roles
• Research plan follows standard structure but is evaluated in the context of career development
• See references/nih-requirements.md for details

SBIR/STTR (R41, R42, R43, R44)

• Milestone-driven aims, not hypotheses: use examples/sbir-specific-aims-template.md
• Reviewed on the five classic criteria (Significance, Investigator(s), Innovation, Approach, Environment), not the RPG Simplified Review Framework
• Commercial potential is scored inside those criteria, in every phase, including Phase I
• Write for a panel where only some reviewers know the field; all of them score
• Restate every result that matters inside the Research Strategy; referenced manuscripts and preprints score nothing
• Phase II, Direct to Phase II, Phase IIB, Fast-Track, and the Commercialization Readiness Pilot require a 12-page Commercialization Plan; a standalone Phase I cannot include one
• Collect letters of interest from potential customers early, not only from academic collaborators
• Phase I runs up to 2 years, so a 24 to 30 month scope is often Phase I plus Phase II rather than a Direct to Phase II
• A Phase I may state an intent to join I-Corps at NIH for customer discovery, which strengthens a later Commercialization Plan
• See references/sbir-sttr-requirements.md, references/commercialization-plan-guide.md, and references/niaid-sample-applications.md

Resubmissions

• Address reviewer concerns point-by-point in an Introduction (1 page)
• Bold changes or mark with change bars
• Keep reviews in a reviews/ directory for reference
• See references/resubmission-guide.md for detailed strategy

Proposal Directory Structure

When creating a new proposal, follow this structure:

Copy code
proposals/{mechanism}/{year}-{short-name}/
├── README.md            # Proposal overview: title, mechanism, PI, deadline, status
├── NOFO.md              # Extracted NOFO requirements with URL link to the original
├── ideas.md             # Early brainstorming, concept development, design decisions
├── research.md          # Background research notes, key references, field context
├── lit-review/          # Literature review materials: search results, annotated bibliographies
├── submission/          # Final submission-ready documents
│   ├── specific-aims.md
│   ├── research-strategy/
│   │   ├── significance.md
│   │   ├── innovation.md
│   │   └── approach.md
│   ├── budget/          # Budget spreadsheets, justification narrative, subaward budgets
│   └── biosketches/     # PI and co-investigator biosketches (SciENcv exports)
├── drafts/              # Working drafts and intermediate versions (not submission-ready)
├── reviews/             # Summary statements, reviewer comments, resubmission notes
└── figures/             # All figures: conceptual diagrams, preliminary data, methods workflows

Additional Resources

Reference Files

For detailed guidance, consult:

• references/nih-requirements.md - NIH mechanisms, page limits, review criteria, deadlines
• references/nsf-requirements.md - NSF mechanisms, formatting, review criteria
• references/research-strategy-guidelines.md - Detailed section-by-section writing guide for NIH research strategy
• references/writing-style-guide.md - Comprehensive scientific writing style guide
• references/tone-guide.md - Narrative tone patterns and voice conventions
• references/budget-justification-guide.md - Budget categories, modular vs detailed, common pitfalls
• references/career-award-guide.md - NSF CAREER award requirements and integration guidance
• references/resubmission-guide.md - Resubmission strategy, Introduction page format, change marking
• references/sbir-sttr-requirements.md - SBIR/STTR program structure, budgets, phase choice, Fast-Track, eligibility, and how small-business review differs
• references/commercialization-plan-guide.md - The 12-page Commercialization Plan, section by section, with the criticisms weak plans draw
• references/niaid-sample-applications.md - Curated index of the NIAID funded sample applications and summary statements, with what to study in each

Example Templates

• examples/specific-aims-template.md - Annotated specific aims page with placeholders (hypothesis-driven; not for SBIR/STTR)
• examples/sbir-specific-aims-template.md - Milestone-driven specific aims for SBIR/STTR, with a contrast table against the generic template
• examples/budget-justification-template.md - Modular and detailed budget formats

Scripts

• scripts/fetch-niaid-samples.sh - Downloads the NIAID small-business sample applications and summary statements, optionally converting them to markdown. The samples are copyrighted and licensed for nonprofit educational use only, so they are fetched on demand rather than bundled. Manifest: scripts/niaid-samples.txt.

Literature Search

Use the fully qualified opencite:opencite skill through the host's skill-invocation mechanism for literature searches to support proposals:

bashCopy code
uvx opencite search "topic" --max 20 -f bibtex -o refs.bib
uvx opencite canonical "field" --max 10