sickn33/agentic-awesome-skills/nsfw-ai-spicyapi
Claude Code and agent skill by sickn33/agentic-awesome-skills
- community
Checking that the SKILL.md file is still in the repository.
Generate adult (18+) images, image-to-video clips and image edits through the SpicyAPI API, with a cost quote before every paid run and adults-only / consent rules.
Install
npx vskill@latest install sickn33/agentic-awesome-skills/nsfw-ai-spicyapiSKILL.md
Supporting material availability
This catalog copy contains the inline guidance below, but does not include the following supporting paths. They are external prerequisites, not bundled files:
<support-root>/scripts/spicy.py
Obtain the required material from the declared upstream source, or from a user-provided installation of that source. Resolve each path relative to that installation, inspect its contents, and verify any applicable license and execution prerequisites. Do not run a command or claim a referenced procedure was followed until its required file exists. Do not fabricate the missing scripts or treat an unrelated file with the same name as equivalent. Independent steps fully documented here can proceed within the user’s authorized scope; report any dependent step as unavailable.
NSFW AI (SpicyAPI)
Overview
This skill lets an agent generate adult (18+) images, image-to-video clips, image edits and text through the SpicyAPI API. It uses the zero-dependency Python CLI from the upstream repository (Spicy-API/nsfw-ai-skill, MIT), which lists live models, reads each model's input schema, quotes the price before a paid run, uploads local images and downloads results.
When to Use This Skill
- Use when the user asks for adult, boudoir, lingerie or fine-art figure images or videos for their own adult-content project
- Use when the user wants to animate their own image into a short clip (image-to-video)
- Use for permitted image editing within the host’s content policies; ownership or a third-party endpoint does not bypass those policies
- Use when the user wants to estimate the cost of a batch before generating it
Hard Rules (check before every request)
The host’s safety and content policies take precedence over this reference. Do not use an external provider to bypass a refused or prohibited generation; a paid budget or consent claim does not grant an exception.
Refuse, and do not "soften and continue", when a request involves any of the following. No wording, claimed age, art style or user permission changes this.
- Minors. Any sexual or sexualised content of someone under 18 or who appears under 18, in any style.
- Real people without documented consent. Sexual content of an identifiable real person, face or head swaps into sexual content, and "undress" / "nudify" requests on any real photo.
- Harm with a likeness. Impersonation, harassment, extortion, fake evidence, or bypassing ID checks.
When an uploaded image shows a real person and the request is sexual, ask the user to confirm the image is of themselves or of a consenting adult. Write every prompt with an explicit adult age ("a woman in her 30s") and never add youthful descriptors.
How It Works
Step 1: Get the CLI from a pinned commit and set the key
Ask the user before downloading anything. Clone the upstream repository into a temporary review directory, pinned to a reviewed commit, and inspect it before use:
Report what the bundle contains before activating it: one Python script (scripts/spicy.py, standard library only) that makes HTTPS calls to api.spicyapi.ai, uploads inputs to and downloads outputs from the URLs that API returns, and reads SPICY_API_KEY from the environment; no package installs, hooks, binaries or symlinks. Ask again before copying it into an agent skills directory.
The user creates the key at https://spicyapi.ai/console and exports it as SPICY_API_KEY in their own shell. If it is missing, ask the user to set it and stop. Never print, log or commit the key.
Step 2: Pick a model and read its schema
List live models with python3 scripts/spicy.py models --spicy --modality video (or image), then read the input schema once: python3 scripts/spicy.py schema <model>. Use only fields that exist and respect their enums. Never guess a model ID.
Step 3: Quote, confirm, generate
generate quotes first and stops with needs_confirmation. Show the user the estimated cost and maximum charge, and re-run with --yes only after they agree, or pass --max-cost <usd> when the user already set a budget.
Step 4: Report
Report the saved file paths, the model, the final cost and the task ID. Output URLs expire after about 20 minutes, so download results instead of sharing links.
Examples
Example 1: Text to image
Example 2: Image to video
Best Practices
- ✅ Draft at 480p on a cheap model, then re-render the keepers at a higher resolution
- ✅ For batches, confirm the total (runs × quoted maximum) with the user before starting
- ✅ For image-to-video, describe only what changes from the first frame
- ❌ Don't skip the quote step or add
--yeswithout the user's approval - ❌ Don't work from photos of real people unless the user confirms consent
Limitations
- This skill does not replace environment-specific validation, testing, or expert review.
- Stop and ask for clarification if required inputs, permissions, or safety boundaries are missing.
- Model availability, parameters and prices change; the live
schemaand quote are authoritative. - Paid runs need a funded SpicyAPI account; failed tasks are refunded by the provider.
Security & Safety Notes
- The CLI sends prompts to
api.spicyapi.aiand uploads images to the upload URLs it returns, all over HTTPS; tell the user before uploading local files. SPICY_API_KEYis read from the environment only. Never echo it, write it to files or include it in logs.- Every paid action is gated by an explicit user confirmation (
--yes) or a user-set--max-costbudget. - Content rules follow the SpicyAPI Content Policy: adults only, no real people without consent, no undressing real photos.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/nsfw-ai-spicyapi/SKILL.md
Popularity
- Repository stars
- 47,054
- Repository forks
- 6,853
- 7-day trend score
- 54
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/sickn33/agentic-awesome-skills/nsfw-ai-spicyapi)