favicon-generator
Generate flat favicons from image prompts, then key out a magenta background and build PNG/ICO/WebP outputs with ImageMagick. Use when you need a reliable favicon workflow.
Install with Tessl CLI
npx tessl i github:Dicklesworthstone/pi_agent_rust --skill favicon-generator82
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Evaluation results
100%
80%Favicon Automation Script for Recipe Platform
Favicon generation workflow scripting
Magenta background in prompt
0%
100%
No magenta in icon
0%
100%
Flat/no gradients style
37%
100%
Centered, fills 75-85%
0%
100%
Bold geometric shapes
0%
100%
Timestamp filename format
0%
100%
1K resolution flag
0%
100%
generate_image.py used
37%
100%
faviconize.sh invoked
0%
100%
ICO output referenced
100%
100%
Apple touch icon referenced
100%
100%
WebP output referenced
0%
100%
1:1 square aspect in prompt
0%
100%
Without context: $0.4368 · 1m 43s · 32 turns · 957 in / 5,535 out tokens
With context: $0.3035 · 1m 10s · 17 turns · 1,032 in / 3,559 out tokens
100%
64%Add Favicon Support to a Web Application
HTML and manifest favicon integration
ICO link tag
50%
100%
32px PNG link tag
41%
100%
Apple touch icon link
41%
100%
Manifest link tag
100%
100%
Manifest 192px PNG
35%
100%
Manifest 512px PNG
35%
100%
Manifest WebP icon
0%
100%
WebP icon src
0%
100%
Without context: $0.2364 · 51s · 14 turns · 20 in / 2,982 out tokens
With context: $0.2506 · 50s · 19 turns · 271 in / 2,758 out tokens
100%
72%Convert Third-Party Icon Assets to Favicon Formats
Custom background key color and FUZZ tuning
faviconize.sh used for shield
0%
100%
faviconize.sh used for leaf
0%
100%
FAVICON_KEY_COLOR for shield
0%
100%
FAVICON_KEY_COLOR for leaf
0%
100%
FAVICON_FUZZ for leaf
0%
100%
Shield output base name
80%
100%
Leaf output base name
75%
100%
Env vars as prefixes
0%
100%
Input files as arguments
100%
100%
Skills path referenced
100%
100%
Without context: $0.4327 · 2m 3s · 31 turns · 164 in / 5,699 out tokens
With context: $0.2087 · 41s · 14 turns · 20 in / 2,069 out tokens
- Evaluated
- Agent
- Claude Code
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.