Grant Mock Reviewer

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

Grant Mock Reviewer

A simulated NIH study section reviewer that provides structured, rigorous critique of grant proposals using the official NIH scoring criteria and methodology.

Quick Check

Use this command to verify that the packaged script entry point can be parsed before deeper execution.

python -m py_compile scripts/main.py

Audit-Ready Commands

Use these concrete commands for validation. They are intentionally self-contained and avoid placeholder paths.

python -m py_compile scripts/main.py
python scripts/main.py --help
python scripts/main.py -h
python scripts/main.py --help

When to Use

• Use this skill when the task needs Simulates NIH study section peer review for grant proposals. Triggers.
• Use this skill for academic writing tasks that require explicit assumptions, bounded scope, and a reproducible output format.
• Use this skill when you need a documented fallback path for missing inputs, execution errors, or partial evidence.

Workflow

1. Confirm the user objective, required inputs, and non-negotiable constraints before doing detailed work.
2. Validate that the request matches the documented scope and stop early if the task would require unsupported assumptions.
3. Use the packaged script path or the documented reasoning path with only the inputs that are actually available.
4. Return a structured result that separates assumptions, deliverables, risks, and unresolved items.
5. If execution fails or inputs are incomplete, switch to the fallback path and state exactly what blocked full completion.

Capabilities

1. NIH Scoring Rubric Application: Official 1-9 scale scoring across all 5 criteria
2. Weakness Identification: Systematic detection of common proposal flaws
3. Critique Generation: Structured written critiques for each review criterion
4. Summary Statement: Complete mock Summary Statement output
5. Revision Guidance: Prioritized, actionable recommendations for improvement

Usage

Command Line

# Full mock review with Summary Statement
python3 scripts/main.py --input proposal.pdf --format pdf --output review.md

# Review Specific Aims only
python3 scripts/main.py --input aims.pdf --section aims --output aims_review.md

# Targeted review (specific criterion focus)
python3 scripts/main.py --input proposal.pdf --focus approach --output approach_critique.md

# Generate NIH-style scores only
python3 scripts/main.py --input proposal.pdf --scores-only --output scores.json

# Compare before/after revision
python3 scripts/main.py --original original.pdf --revised revised.pdf --compare

As Library

from scripts.main import GrantMockReviewer

reviewer = GrantMockReviewer()
result = reviewer.review(
    proposal_text=proposal_content,
    grant_type="R01",
    section="full"
)
print(result.summary_statement)
print(result.scores)

Parameters

NIH Scoring System

Overall Impact Score (1-9)

The single most important score reflecting the likelihood of the project to exert a sustained, powerful influence on the research field.

Individual Criteria (1-9 each)

1. Significance: Does the project address an important problem? Will scientific knowledge be advanced?
2. Investigator(s): Are the PIs well-suited? Adequate experience and training?
3. Innovation: Does it challenge current paradigms? Novel concepts, approaches, methods?
4. Approach: Sound research design? Appropriate methods? Adequate controls? Address pitfalls?
5. Environment: Adequate institutional support? Scientific environment conducive to success?

Score Interpretation

• 1-3 (High Priority): Compelling, well-developed proposals with strong approach
• 4-5 (Medium Priority): Good proposals with some weaknesses
• 6-9 (Low Priority): Significant weaknesses that diminish enthusiasm

Review Output Format

1. Score Summary

Overall Impact: [Score] - [Descriptor]

Criterion Scores:
- Significance: [Score]
- Investigator(s): [Score]
- Innovation: [Score]
- Approach: [Score]
- Environment: [Score]

2. Strengths

Bullet-point list of major strengths by criterion

3. Weaknesses

Bullet-point list of major weaknesses by criterion

4. Detailed Critique

Paragraph-form critique for each criterion following NIH style

5. Summary Statement

Complete narrative synthesis of the review

6. Revision Recommendations

Prioritized, actionable suggestions for improvement

Common Weaknesses Detected

Significance

• Insufficient justification for the research problem
• Incremental rather than transformative impact
• Unclear connection to human health/disease
• Overstatement of clinical significance without evidence

Investigator

• Lack of relevant expertise for proposed aims
• Insufficient track record in key methodologies
• PI overcommitted (excessive effort on other grants)
• Missing key collaborator expertise

Innovation

• Straightforward extension of published work
• Methods are standard rather than novel
• No challenging of existing paradigms
• Incremental rather than breakthrough potential

Approach

• Aims too ambitious for timeframe
• Insufficient preliminary data
• Inadequate experimental controls
• No discussion of pitfalls and alternatives
• Statistical analysis plan missing or inadequate
• Sample size/power calculations absent

Environment

• Inadequate institutional resources
• Missing core facility access
• Lack of relevant equipment
• Insufficient collaborative environment

Technical Difficulty

High - Requires deep understanding of NIH peer review processes, ability to apply standardized scoring rubrics consistently, and generation of clinically/scientifically accurate critique across diverse research domains.

Review Required: Human verification recommended before deployment in production settings.

References

• references/nih_scoring_rubric.md - Complete NIH scoring guidelines
• references/review_criteria_explained.md - Detailed criterion descriptions
• references/common_weaknesses_catalog.md - Database of typical proposal flaws
• references/summary_statement_templates.md - NIH-style statement templates
• references/score_calibration_guide.md - Score assignment guidelines

Best Practices for Users

1. Provide Complete Proposals: The tool works best with full Research Strategy sections
2. Include Preliminary Data: Approach critique depends on feasibility evidence
3. Review Multiple Times: Use iteratively as you revise
4. Compare Versions: Track improvement between drafts
5. Consider Multiple Perspectives: Supplement with human reviewer feedback

Limitations

1. Cannot access external literature to verify claims
2. May not capture domain-specific methodological nuances
3. Scoring is simulated and may not match actual study section scores
4. Best used as preparatory tool, not replacement for human review

Version

1.0.0 - Initial release with NIH R01/R21/R03 support

Risk Assessment

Security Checklist

• No hardcoded credentials or API keys
• No unauthorized file system access (../)
• Output does not expose sensitive information
• Prompt injection protections in place
• Input file paths validated (no ../ traversal)
• Output directory restricted to workspace
• Script execution in sandboxed environment
• Error messages sanitized (no stack traces exposed)
• Dependencies audited