Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add avenoxai/avenoxskills --skill avenox-thumbnailgit clone --depth 1 https://github.com/avenoxai/avenoxskillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/avenoxai/avenoxskills/avenox-thumbnail)<a href="https://agentmods.dev/skills/avenoxai/avenoxskills/avenox-thumbnail"><img src="https://agentmods.dev/badge/skills/avenoxai/avenoxskills/avenox-thumbnail/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/avenoxai/avenoxskills/avenox-thumbnail"><img src="https://agentmods.dev/badge/skills/avenoxai/avenoxskills/avenox-thumbnail.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00102 | $0.01289 |
| Opus 5 | $0.00051 | $0.00645 |
| Sonnet 5 | $0.00020 | $0.00258 |
| Haiku 4.5 | $0.00010 | $0.00129 |
Grade A, and why
avenox-thumbnail scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Thumbnail Factory
A systematic pipeline for producing scroll-stopping YouTube thumbnails for a mascot-driven channel. It encodes the formula, the working prompt recipe, the hook-pattern playbook, and the hard safety rules learned in production.
Built for anonymous / faceless channels: the mascot is the face, so the covers stay consistent without ever putting a person on screen.
Anonymity rule: never put the creator's real name, personal handle, or likeness on a thumbnail. If the channel is anonymous, the mascot is the identity — breaking that once is not reversible.
Configure before first use
| Setting | Where | Example |
|---|---|---|
| Mascot reference image | assets/mascot-ref.png |
see assets/README.md |
| Mascot on-model contract | references/character-and-safety.md |
fill in the contract block |
| Hook language | prompt (HOOK_LANG) |
Turkish, English, … |
| Output dir | $THUMB_OUT |
~/Desktop/thumbnails |
The formula
Every thumbnail = 4 ingredients, nothing more (3-element rule; blink-test first — if it doesn't read in a quarter second at sidebar size, it failed):
- The mascot — on-model, expressive pose that fits the video's emotion.
- One 2-4 word hook — MASSIVE bold sans-serif caps, white + ONE accent word, thick black outline. Never duplicate the video title; open a curiosity gap that the title and video close.
- Relevant tool logo(s) — so the topic is legible in one glance.
- High-contrast modern background — dark + one strong accent color (red = urgency, green = success, blue = tech, purple = automation, gold = money). Bright, saturated, clean.
Design levers (apply, don't just decorate): curiosity gap • negativity/risk • contrast (before/after, ✓/✗) • tangible number or timeframe • authority • one clear focal point • lower-right corner kept clear (YouTube stamps the duration there) • text left / mascot right reads best.
Full hook playbook → references/hook-patterns.md.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 111 lines · 102 tokens per session scan A ceadbc2e0fe3
avenox-thumbnail is a skill published in the GitHub repository avenoxai/avenoxskills (49 stars, last pushed 1mo ago), licensed MIT. It adds 102 tokens to every session and 1,289 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
pptx
Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying…
edu-math-tutorial
A Chinese-language guide for turning a maths problem into a step-by-step teaching video. It explains how to break down the solution, write narration, format equations, and structure scenes.
banner-creator
Create banners using AI image generation. Discuss format/style, generate variations, iterate with user feedback, crop to target ratio. Use when user wants to create a banner, header, hero image, cover image, GitHub banner, Twitter header, or readme banner.
logo-creator
Create logos using AI image generation. Discuss style/ratio, generate variations, iterate with user feedback, crop, remove background, and export as SVG. Use when user wants to create a logo, icon, favicon, brand mark, mascot, emblem, or design a logo.
nanobanana
Generate and edit images using Google Gemini 3 Pro Image (Nano Banana Pro). Supports text-to-image, image editing, various aspect ratios, and high-resolution output (2K/4K). Use when user wants to generate images, create images, use Gemini image generation, or do AI image generation.
gpt-image-gen
MUST read before generating images. Detailed prompt-crafting guide for gpt-image models, covering tool routing (native imagegeneration server tool vs the generateimage tool), prompt structure from subject to background, verbatim text rendering, anti-patterns, and the revisedprompt feedback loop for iteration.