Photo Edit Analysis

Analyze the composition, editing, and post-processing quality of a photograph. Use when a user shares a photo and asks about the edit, exposure, tone, color...

Security Scan

This is an instruction-only photo critique skill whose declared purpose, runtime instructions, and lack of required credentials or installs are consistent and proportionate.

Purpose & Capability

Name and description match the SKILL.md and README: the skill provides composition and edit critique. It requests no binaries, credentials, or config paths that would be unrelated to image analysis.

Instruction Scope

Runtime instructions focus narrowly on composition and post-processing critique and specify output format and tone. They do not direct the agent to read unrelated files, access credentials, or send data to external endpoints.

Install Mechanism

No install spec and no code files — instruction-only. Nothing is downloaded or written to disk by the skill itself, minimizing install-time risk.

Credentials

The skill declares no environment variables, credentials, or sensitive config paths. There are no disproportionate secret or credential requests.

Persistence & Privilege

always is false and the skill is user-invocable. It does not request persistent presence or system-wide privileges.

Assessment

This skill appears internally consistent and safe from the package contents perspective. Before using it, consider privacy of any images you submit (personally identifying details, sensitive contexts). The SKILL.md does not mention external uploads, but the agent platform or model provider may send image content to remote inference services — confirm where image data is processed and logged. If you plan to use this alongside other skills (the README mentions integration with a `photo-captions` skill), review that other skill’s permissions and data flows as well. Avoid submitting highly sensitive images unless you’ve verified the processing and retention policy.

When a user shares a photo for edit or composition feedback, analyze both and deliver honest, specific notes.

What to Analyze

Composition

• Framing & subject placement — rule of thirds, leading lines, visual balance, negative space
• Foreground/mid-ground/background relationship — does the image have depth and layering?
• Point of entry — where does the eye land first, and where does it travel?
• Horizon line — level? intentionally tilted? does it work?
• Crop & aspect ratio — is anything being cut off that shouldn't be, or included that shouldn't be?
• Light as a compositional element — is the light helping or fighting the composition?

Editing & Post-Processing

• White balance & color grade — warm/cool bias, color cast, whether it suits the subject and mood
• Contrast & tonal curve — shadow crush, highlight blow, midtone separation, overall tonal range
• Saturation & color rendering — oversaturated? flat? are individual color channels serving the image?
• Shadow & highlight handling — is detail retained in both extremes? any clipping?
• Clarity & texture — is sharpness natural or over-processed? does it suit the mood?
• Grain — consistent, natural, appropriate for the stock/ISO? or noise masquerading as grain?
• Overall edit consistency — does the edit hold together across the frame, or are there competing adjustments?

Output Format

Two sections: Composition then Edit. Lead each with what's working, then what isn't. Be specific — name actual zones, colors, lines, or tonal regions. Don't hedge everything.

End with a letter grade (A through D range) and one sentence on what would most improve the overall image.

Keep it to 200-300 words total. Editorial feedback, not a technical report.

Tone

Talk like a photo editor giving notes to a photographer they respect. Direct, honest, no filler. Don't say "overall this is great" if it isn't. Don't say "interesting choice" as a euphemism for "this doesn't work."

Film-specific: account for the stock's characteristics. Portra 400 overexposed by a stop is a deliberate choice, not a mistake. Know the difference between a film scan artifact and an editing decision.

Guidelines

• Never make the gear or film stock the subject of the critique. The image is the subject.
• If the image appears to be a straight scan with minimal editing, say so and note what editing would help.
• If the edit or composition is strong, say so clearly. A short positive critique is still useful.
• Don't repeat observations from the caption generation. The analysis should stand on its own.