Scoring OptimizerSAFE

An agent skill published by revfactory.

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: scoring-optimizer
description: "Methodology for analyzing government funding program evaluation scoring tables and developing high-score strategies. Use this skill for 'scoring analysis', 'high-score strategy', 'evaluation criteria optimization', 'review preparation', 'document/presentation evaluation preparation', and other government project evaluation preparation tasks. Note: contacting reviewers and conducting actual presentations are outside the scope of this skill."
---

# Scoring Optimizer — Government Funding Evaluation Scoring Strategy

A skill that enhances evaluation preparation for the submission-reviewer, tech-writer, and biz-writer.

## Target Agents

- **submission-reviewer** — Optimizes the proposal for maximum scoring
- **tech-writer** — Aligns technical content with high-weight criteria
- **biz-writer** — Aligns business content with high-weight criteria

## Scoring Analysis Framework

### Step 1: Weight Distribution Analysis
```
Map evaluation criteria by weight:
- High weight (>15%): Must address thoroughly, allocate proportional content
- Medium weight (10-15%): Address completely, include evidence
- Low weight (<10%): Address adequately, do not over-invest
```

### Step 2: Difficulty-Impact Matrix
```
| Criteria | Weight | Difficulty | Strategy |
|----------|--------|-----------|----------|
| High weight + Easy | Priority 1 | Full score target |
| High weight + Hard | Priority 2 | Maximize with best evidence |
| Low weight + Easy | Priority 3 | Quick win, minimal effort |
| Low weight + Hard | Priority 4 | Adequate coverage only |
```

### Step 3: Evidence Mapping
```
For each criterion:
1. What specific evidence demonstrates this?
2. Is it quantitative or qualitative?
3. Where in the proposal does it appear?
4. Can we strengthen the evidence?
```

## Common Evaluation Categories

### Technical Excellence (typically 30-40%)
```
High-score tactics:
- Clear, measurable technical objectives
- Comparison tables vs. existing technology
- Technology readiness level (TRL) progression
- Risk mitigation plans
- Team expertise matching technical needs
```

### Business Feasibility (typically 20-30%)
```
High-score tactics:
- Market size with credible sources
- Specific commercialization timeline
- Revenue projections with assumptions
- IP strategy
- Partnership/customer letters of intent
```

### Implementation Plan (typically 15-20%)
```
High-score tactics:
- Detailed milestone-based timeline
- Clear role assignments
- Resource allocation justification
- Quality assurance plan
```

### Budget Appropriateness (typically 10-15%)
```
High-score tactics:
- Line items match technical plan exactly
- Government standard rates applied
- Clear justification for each cost item
- Reasonable proportion across categories
```

## Presentation Preparation (if applicable)
```
Time management:
- Intro: 10% of time
- Problem/Market: 15%
- Technology: 30%
- Business: 20%
- Team/Plan: 15%
- Q&A prep: 10%

Key rules:
- One key message per slide
- Reviewers evaluate both content and presentation quality
- Anticipate and prepare for tough questions
- Have backup slides for detailed data
```

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.