Agent skill · research science · aradotso
k-dense-byok-ai-scientist
k-dense-byok-ai-scientist — an agent skill for coding agents.
Why this skill is useful
Adds executable scripts for setting up a local AI research assistant with integration to multiple scientific databases and models.
What it needs
Requires git, node, python installed locally. Requires modal, openrouter, parallel account access. About 5k tokens when loaded. Last updated 2026-07-09. 70 stars on the source repository.
What this skill does
bash git clone https://github.com/K-Dense-AI/k-dense-byok.git cd k-dense-byok env Required OPENROUTERAPIKEY=$OPENROUTERAPIKEY Optional — enables web search PARALLELAPIKEY=$PARALLELAPIKEY Internal service wiring — leave as-is GOOGLEGEMINIBASEURL=http://localhost:4000 GEMINIAPIKEY=sk-litellm-local Default model for Kady (changeable in UI dropdown) DEFAULTAGENTMODEL=openrouter/google/gemini-3.1-pro-preview Optional — Modal remote compute MODALTOKENID=$MODALTOKENID MODALTOKENSECRET=$MODALTOKENSECRET bash chmod +x start.sh ./start.sh k-dense-byok/ ├── start.sh # One-command launcher ├── server.py # FastAPI backend ├── kadyagent/ │ ├── .env # API keys │ ├── agent.py # Main Kady agent definition │ └── tools/ # Web search, delegation, file ops ├── web/ # Next.js frontend └── sandbox/ # Local file workspace (auto-created) python kadyagent/agent.py (illustrative structure) from anthropic import Anthropic client = Anthropic() Kady decides: answer directly OR delegate to an expert Experts have access to 170+ scientific skills typescript // web/src/lib/api.ts — example frontend API call pattern const response = await fetch('http://localhost:8000/chat', { method: 'POST', headers: { 'Content-Type': 'application/json' }, body: JSON.stringify({ message: userMessage, sessionid: sessionId, model: selectedModel, // e.g. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aradotso/k-dense-byok-ai-scientist