Skip to content

deapi-ai/skills

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 

Repository files navigation

deAPI Skills

Agent skills for deAPI — let AI agents generate images, speech, music and video, transcribe audio, extract text from images, and more.

Compatible with the agentskills.io open standard: Hermes Agent, Claude Code, Cursor, Goose, OpenCode, Codex, GitHub Copilot, Windsurf and 40+ other clients. Up to 20x cheaper inference via decentralized GPU network.

Skills

Skill Description
deapi All deAPI modalities through one portable skill: text-to-image (FLUX.2 Klein), image editing, TTS, transcription (YouTube/X/Twitch links or files), OCR, background removal, upscaling, music, video, embeddings. Python (stdlib only), live model discovery — no hardcoded slugs.
deapi-audio Audio only, bash + curl/jq: text-to-speech, voice cloning, voice design, audio transcription.

Pick deapi for full coverage and portability (works in Hermes with automatic API-key handling); pick deapi-audio if you want zero-Python bash scripts for audio tasks.

Setup

1. Get your API key

Sign up at deapi.ai (free $5 credit, no card required) and create a key (format dpn-sk-...).

2. Set the key

export DEAPI_API_KEY="dpn-sk-..."

(Hermes Agent instead asks for the key on first use and stores it in ~/.hermes/.env.)

3. Install

Hermes Agent:

mkdir -p ~/.hermes/skills/creative
git clone https://github.com/deapi-ai/skills.git
cp -r skills/deapi ~/.hermes/skills/creative/deapi

Claude Code:

mkdir -p ~/.claude/skills
git clone https://github.com/deapi-ai/skills.git
cp -r skills/deapi ~/.claude/skills/       # and/or skills/deapi-audio

Cursor / Goose / OpenCode / Codex / VS Code Copilot (shared convention):

mkdir -p ~/.agents/skills
cp -r skills/deapi ~/.agents/skills/

Or via skills.sh: npx skills add deapi-ai/skills

4. Use

Ask your agent naturally — "generate an image of a red fox", "transcribe this YouTube video", "read the text in scan.png". Direct CLI also works:

python3 deapi/scripts/deapi.py image --prompt "a red fox, studio light" --output fox.png
python3 deapi/scripts/deapi.py stt --url "https://www.youtube.com/watch?v=..."
python3 deapi/scripts/deapi.py models --type image
bash deapi-audio/scripts/text-to-speech.sh --text "Hello world" --voice bf_emma

Skill Format

Each skill follows the agentskills.io standard:

deapi/
├── SKILL.md              # YAML frontmatter (name, description, metadata) + docs
├── scripts/              # executable helpers
└── references/           # API reference, model discovery guide

Related

License

MIT

About

Specialized skills for deAPI service

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors