Award ApplicationSAFE

Skills for Claude Code CLI such as full stack dev Cloudflare, React, Tailwind v4, and AI integrations.

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: award-application
description: "Write compelling award submissions, grant applications, and competition entries. Maps achievements to selection criteria using evidence-based narratives. Use whenever the user is applying for a business award (Telstra, chamber of commerce, industry awards), a grant, a competition, or any submission demonstrating merit against defined criteria."
---

# Award Application Writer

Produces award submissions and grant applications that address every criterion with quantified evidence. The goal is a document where every paragraph earns points on the scorecard.

## Process

### Step 1: Gather the brief

Ask the user for:

- Award/grant name and organiser
- Selection criteria (the exact list, verbatim if possible)
- Word limits (per criterion and/or total)
- Judging rubric or weighting (if publicly available)
- Category entered (if multiple categories exist)
- Whether this is a first submission or a resubmission
- Any specific achievements, metrics, or stories they want included

If the user has a URL for the award, fetch the criteria page. Many awards publish judging guides or past winner profiles — these are gold for understanding what evaluators value.

### Step 2: Map achievements to criteria

Create a working table before writing anything:

| Criterion | Key achievement | Evidence/metric | Story or example |
|-----------|----------------|-----------------|------------------|
| Innovation | AI workflow automation | Report time: 3 days to 4 hours | Staff training program, 12 people |
| Growth | Revenue increase | 40% YoY, $X to $Y | New service line launched Q2 |
| Community | Pro bono program | 200 hours, 15 local orgs | Bushfire recovery site builds |

Every criterion must have at least one entry. A blank row means a missing section in the submission — and missing sections lose marks or trigger automatic rejection.

If the user cannot provide evidence for a criterion, flag it explicitly. Better to know the gap now than to submit vague filler.

### Step 3: Write the submission

Address criteria in the exact order they appear on the application form. Judges often score sequentially — don't make them hunt for your answer to criterion 3 buried in the criterion 1 section.

For each criterion, use the STAR structure:

- **Situation**: Brief context (1-2 sentences). What was the starting point or challenge?
- **Task**: What needed to happen? What was the objective?
- **Action**: What specifically did you/the business do? Be concrete.
- **Result**: What changed? Quantify the outcome.

Keep each criterion response self-contained. A judge reading only that section should understand the achievement without needing context from other sections.

### Step 4: Review against limits

- Check word counts per section and total
- Verify every criterion is addressed
- Confirm all claims have supporting evidence
- Read the opening line of each section — does it lead with impact?

## Writing Approach

### Lead with impact, not chronology

The first sentence of every section should be your strongest claim.

**Wrong approach:**
> Founded in 2018, Acme Digital began as a two-person consultancy. Over the following years, we grew steadily, adding new team members and services. In 2025, we achieved significant growth.

**Right approach:**
> Acme Digital grew revenue 40% in 12 months ($850K to $1.19M) while maintaining a 94% client retention rate. This growth came from a deliberate shift into AI-powered automation services, launched in Q1 2025.

Chronology can appear in the body as context, but never as the opening.

### Quantify everything possible

Judges compare applicants. Numbers make comparison easy and your claims credible.

| Vague | Quantified |
|-------|-----------|
| "Significant growth" | "Revenue increased 40% ($850K to $1.19M)" |
| "Many clients" | "127 active clients across 3 states" |
| "Improved efficiency" | "Reduced report generation from 3 days to 4 hours" |
| "Community involvement" | "Donated 200 hours of pro bono work to 15 local organisations" |
| "Award-winning team" | "Team of 8, including 2 certified Google Partners and 1 Shopify Expert" |

If exact numbers are not available, use defensible approximations with qualifiers: "approximately", "more than", "over the past 12 months".

### Show, don't tell

Remove adjectives. Replace them with evidence.

| Telling | Showing |
|---------|---------|
| "We're innovative and forward-thinking" | "Trained 12 staff in AI tools, reducing average report time from 3 days to 4 hours" |
| "We deliver exceptional customer service" | "Net Promoter Score of 72, with 94% client retention over 3 years" |
| "We're passionate about our community" | "Built 6 pro bono websites for Hunter Valley bushfire-affected businesses in 2025" |

### Respect word limits absolutely

Many awards auto-disqualify entries that exceed word limits. If the limit is 500 words, submit 480-500. Using significantly fewer words than allowed leaves points on the table.

When tight on words:
- Cut setup/context sentences first (judges know the industry)
- Combine STAR elements where natural
- Remove hedging language ("we believe", "we feel that", "it could be said")
- One strong example beats three weak ones

## What Different Awards Look For

### Business awards (Telstra, local chamber, BEC)

- Growth trajectory with numbers
- Innovation in products, services, or operations
- Leadership and team development
- Community contribution and social impact
- Resilience and adaptability (especially post-COVID, post-disaster)

### Industry/professional awards

- Technical excellence and methodology
- Client outcomes with measurable results
- Thought leadership (publications, speaking, mentoring)
- Industry contribution beyond your own business

### Grants

- Clear problem statement (what needs fixing and why it matters)
- Feasibility (can you actually deliver this?)
- Budget justification (every line item tied to an activity)
- Measurable outcomes (how will you kn

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.