Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/timoncool/ACE-Step-Studionpx agentmods add skills/timoncool/ace-step-studio/acestep-thumbnailWrote 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/timoncool/ace-step-studio/acestep-thumbnail)<a href="https://agentmods.dev/skills/timoncool/ace-step-studio/acestep-thumbnail"><img src="https://agentmods.dev/badge/skills/timoncool/ace-step-studio/acestep-thumbnail.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 34 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00045 | $0.01260 |
| Opus 5 | $0.00023 | $0.00630 |
| Sonnet 5 | $0.00009 | $0.00252 |
| Haiku 4.5 | $0.00005 | $0.00126 |
Grade A, and why
acestep-thumbnail scanned grade A with 1 finding 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 8d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- curl, jq, base64 (or python3) How it starts
The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Thumbnail Generation Skill
Generate song cover/thumbnail images using Google Gemini's image generation API. Output images can be used directly as MV backgrounds with the acestep-simplemv skill.
API Key Setup Guide
Before generating, you MUST check whether the user's API key is configured. Run the following command to check:
cd "{project_root}/{.claude or .codex}/skills/acestep-thumbnail/" && bash ./scripts/acestep-thumbnail.sh config --check-key
This command only reports whether the API key is set or empty — it does NOT print the actual key value. NEVER read or display the user's API key content. Do not use config --get on key fields or read config.json directly. The config --list command is safe — it automatically masks API keys as *** in output.
If the command reports the key is empty, you MUST stop and guide the user to configure it before proceeding. Do NOT attempt generation without a valid key — it will fail.
Use AskUserQuestion to ask the user to provide their API key, with the following guidance:
- Tell the user the Gemini API key is not configured and image generation cannot proceed without it.
- Provide instructions on where to get a key:
- Google AI Studio: Get a API key at https://aistudio.google.com/apikey — requires a Google account.
- Once the user provides the key, configure it using:
cd "{project_root}/{.claude or .codex}/skills/acestep-thumbnail/" && bash ./scripts/acestep-thumbnail.sh config --set api_key <KEY> - After configuring, re-run
config --check-keyto verify the key is set before proceeding.
If the API key is already configured, proceed directly to generation without asking.
Quick Start
# 1. cd to this skill's directory
cd {project_root}/{.claude or .codex}/skills/acestep-thumbnail/
# 2. Configure API key
./scripts/acestep-thumbnail.sh config --set api_key <YOUR_GEMINI_KEY>
# 3. Generate thumbnail
./scripts/acestep-thumbnail.sh generate --prompt "Cherry blossoms at night with moonlight"
# 4. Output saved to: {project_root}/acestep_output/<timestamp>_thumbnail.png
What ships with it
2 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.
- 8d ago First seen · 124 lines · 45 tokens per session scan A 1a92f190c416
acestep-thumbnail is a skill published in the GitHub repository timoncool/ACE-Step-Studio (337 stars, last pushed today), licensed MIT. It adds 45 tokens to every session and 1,260 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
claude-music
Music production suite using ACE-Step 1.5 via Python API. Routes /music commands for generation, cover, repaint, compose, analyze, export, enhance, random, and LoRA training. 50+ languages, up to 10-minute tracks, 48kHz stereo.
claude-music-cover
Creates cover versions and style transfers of existing songs using ACE-Step 1.5. Takes a reference audio file and generates a new version with different style, genre, or vocal characteristics while preserving musical structure.
claude-music-repaint
Edits specific sections of a song using ACE-Step 1.5's repaint mode. Fixes a bad chorus, changes instruments in a section, adds or removes vocals, or regenerates any time range while keeping the rest intact.
claude-music-enhance
Post-processing for generated music. Default path is FFmpeg loudness normalization (fast, CPU-only). Escape hatches: AI vocal denoise for artifacts, stem separation for surgical edits. Both escape-hatch paths reuse tooling from the claude-video skill.
claude-music-export
Exports music for specific platforms with optimal format, loudness, and metadata. Supports Spotify, Apple Music, YouTube, TikTok, podcast, CD, streaming, and archive formats. Uses FFmpeg for loudness normalization, format conversion, and metadata tagging.
claude-music-lora
LoRA and LoKr fine-tuning for ACE-Step 1.5. Trains custom styles from 3-10 songs, manages trained adapters, and applies them during generation. Uses ACE-Step's built-in training pipeline.