mirror of
https://github.com/peterlodri-sec/dyad-ultra-auditor.git
synced 2026-11-21 23:29:58 +01:00
Upstream pull mirror of https://github.com/peterlodri-sec/dyad-ultra-auditor.git
- Python 97.4%
- Dockerfile 2.6%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
| .dockerignore | ||
| .env.example | ||
| .gitignore | ||
| CLAUDE.md | ||
| docker-compose.yml | ||
| Dockerfile | ||
| helpers.py | ||
| main.py | ||
| README.md | ||
| reasoners.py | ||
| requirements.txt | ||
dyad-ultra-auditor
A AgentField agent created with af init.
Quick Start
-
Install dependencies:
pip install -r requirements.txt -
Start the AgentField server (in another terminal):
af server -
Run the agent:
python main.pyThe agent will auto-discover an available port and register with AgentField.
Test the Agent
Echo Reasoner (No AI Required)
curl -X POST http://localhost:8080/api/v1/execute/dyad-ultra-auditor.demo_echo \
-H "Content-Type: application/json" \
-d '{"input": {"message": "Hello World"}}'
Enable AI Features
- Uncomment the
ai_configinmain.py - Set your API key:
export OPENAI_API_KEY=sk-...- Or use
ANTHROPIC_API_KEY,GOOGLE_API_KEY, etc. for other providers - LiteLLM auto-detects the provider from the model name
- Or use
- Uncomment the
analyze_sentimentreasoner inreasoners.py - Restart the agent
curl -X POST http://localhost:8080/api/v1/execute/dyad-ultra-auditor.demo_analyze_sentiment \
-H "Content-Type: application/json" \
-d '{"input": {"text": "I love this product!"}}'
Project Structure
main.py- Agent configuration and startupreasoners.py- Reasoner definitions (AI-powered functions)requirements.txt- Python dependencies.env.example- Environment variable examples
Learn More
Created by Lodri Péter (267054113+peterlodri-sec@users.noreply.github.com) on 2026-08-17 08:52:31 CEST