Run Rivaro Locally
Local installs of Rivaro (runtime guard, local dashboard, sample policies) are currently set up together with our team rather than self-serve. We walk every new workspace through setup directly — from connecting your first agent to verifying that runtime enforcement is working — so you get a working deployment without debugging an install on your own.
To get started, contact us and we'll schedule a guided onboarding session.
In the meantime:
- Use hosted Rivaro — create or join a workspace in Rivaro Cloud.
What gets detected
Rivaro scans every request and response for:
| Category | Examples |
|---|---|
| Credentials & secrets | API keys, SSH keys, tokens, passwords |
| PII | Emails, phone numbers, SSNs, addresses |
| Prompt injection | Jailbreak attempts, instruction override |
| Dangerous tool calls | File access, shell commands, network requests |
| Toxic content | Harmful, offensive, or inappropriate output |
| Data exfiltration | Sensitive data leaving to external endpoints |
What's next
- Agent Frameworks — LangChain, CrewAI, Vercel AI SDK, AutoGen integration guides
- Configuration Guide — AppContexts, detection keys, rate limits, and organization setup
- Enforcement & Policies — Configure what happens when violations are detected
- Understanding Detections — What Rivaro scans for and how detections are classified
- Provider Guides — Detailed integration for OpenAI, Anthropic, Azure, Bedrock, Vertex AI
- MCP Governance — Govern MCP tool invocations