VM0Agent Skills
Tavily
AI search API for RAG applications
Tavily is an AI search API designed for retrieval-augmented generation (RAG) with real-time web search and source citations.
Required Environment
| Name | Type | Description |
|---|---|---|
TAVILY_API_KEY | secret | API key from Tavily Dashboard |
Configuration
version: "1.0"
agents:
my-agent:
framework: claude-code
skills:
- https://github.com/vm0-ai/vm0-skills/tree/main/tavilyRun
Store your secret on the platform (recommended, one-time setup):
vm0 secret set TAVILY_API_KEY your-tavily-api-keyThen run your agent - secret is automatically loaded:
vm0 run my-agent "search for latest news"For CI/CD or temporary overrides, pass secrets at runtime: --secrets TAVILY_API_KEY=value. See Environment Variables for details.
Example Instructions
# RAG Search Agent
You search the web for RAG context.
## Workflow
1. Receive search query
2. Execute Tavily search
3. Extract relevant content
4. Return with sources
## Features
- Real-time web data
- Source citations# Research Agent
You conduct web research.
## Workflow
1. Receive research topic
2. Search multiple queries
3. Synthesize findings
4. Generate report
## Search Modes
- Basic search, deep research