CtrlK
BlogDocsLog inGet started
Tessl Logo

nano-banana-pro

Generate/edit images with Nano Banana Pro (Gemini 3 Pro Image). Use for image create/modify requests incl. edits. Supports text-to-image + image-to-image; 1K/2K/4K; use --input-image.

84

3.12x
Quality

76%

Does it follow best practices?

Impact

100%

3.12x

Average score across 3 eval scenarios

SecuritybySnyk

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./tests/ext_conformance/artifacts/agents-mikeastock/skills/nano-banana-pro/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A lean, highly actionable skill: every command is executable and matches the bundled script, the draft-to-final workflow is sensible, and error handling is mapped to concrete fixes. The only meaningful gaps are a slightly redundant Examples section and an iteration loop whose feedback checkpoint is implicit.

DimensionReasoningScore

Conciseness

The body is efficient and assumes competence — no space is spent explaining what image generation is, and sections like the resolution mapping table and failure fixes are pure signal. It falls short of 5 due to minor redundancy: the terminal "Examples" section repeats the Usage commands nearly verbatim, and the Filename section uses four examples where one or two would do.

4 / 5

Actionability

Commands are fully executable and copy-paste ready — `uv run {baseDir}/scripts/generate_image.py --prompt ... --filename ... --input-image ... --resolution 1K|2K|4K` — and every documented flag matches the actual script's argparse interface (verified in scripts/generate_image.py). Preflight checks ("command -v uv", "test -n \"$GEMINI_API_KEY\"", "test -f") and error-message-to-fix mappings are concrete and cover the common cases, matching the anchor-5 example.

5 / 5

Workflow Clarity

The "Default Workflow (draft → iterate → final)" gives a clear, well-reasoned sequence, and the Preflight + Common Failures sections provide pre-checks and error-recovery guidance. It stops short of 5 because there is no explicit validation checkpoint on the draft output (the loop relies on the implicit "until you're happy" and the blanket "Do not read the image back" rule, leaving the iteration feedback mechanism undefined).

4 / 5

Progressive Disclosure

This is a simple, single-purpose skill whose implementation is correctly externalized to one real bundle file (scripts/generate_image.py), referenced one level deep with well-organized sections (Usage, Workflow, Resolution, API Key, Failures, Filename, Editing, Templates, Output). Nothing that belongs in a separate reference file is inlined, matching the simple-skill exception and the anchor-5 structure bar.

5 / 5

Total

18

/

20

Passed

Description

67%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A compact, third-person description that explicitly covers both what the skill does and when to use it, with concrete capabilities and parameters. Its main weakness is trigger-term coverage: it relies on product names and terse phrasing rather than the natural synonyms (photo, picture, draw, render) users are likely to say.

Suggestions

Expand the 'Use when...' clause with concrete trigger phrases, e.g. "Use when the user asks to create, generate, draw, or edit images, photos, or pictures".

Add natural synonyms and file extensions (photo, picture, render, .png/.jpg) to improve trigger-term coverage.

Spell out abbreviations ("incl.", "+") and use full sentences so the description reads as comprehensive rather than telegraphic.

DimensionReasoningScore

Specificity

The description lists several concrete capabilities — "Generate/edit images", "Supports text-to-image + image-to-image; 1K/2K/4K" — plus a concrete parameter ("use --input-image"). It matches anchor 4 (several specific actions, minor gaps) rather than 5 because the compressed telegraphic style ("incl.", "+") leaves coverage slightly terse rather than comprehensive.

4 / 5

Completeness

Both parts are present: "what" ("Generate/edit images with Nano Banana Pro... Supports text-to-image + image-to-image; 1K/2K/4K") and an explicit "when" ("Use for image create/modify requests incl. edits"). It is not a 5 because the 'when' clause is brief and lacks the concrete trigger phrases of the anchor-5 example, and not a 3 because 'when' is explicitly stated rather than implied.

4 / 5

Trigger Term Quality

It includes relevant terms like "image create/modify requests", "edits", and "text-to-image", but misses common natural variations users would say — "photo", "picture", "draw", "render", "make an image" — matching anchor 3 (some relevant keywords, missing common synonyms) rather than 4's fuller keyword coverage.

3 / 5

Distinctiveness Conflict Risk

Naming "Nano Banana Pro (Gemini 3 Pro Image)" carves out a clear product niche with distinct triggers, but the generic "Generate/edit images" framing leaves minor overlap risk with other image-generation/editing skills, matching anchor 4 rather than 5's minimal-conflict bar.

4 / 5

Total

15

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
Dicklesworthstone/pi_agent_rust
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.