Find every agent
Scan your code for every agent, model, and tool, including the ones nobody registered. Read-only, in less than 10 minutes.
Test before real work
Each agent faces adversarial tests such as prompt injection, and gets a resilience score you can track.
Guard it and keep the record
Guardrails check what goes into and out of each agent, with live traces of what it does. Add the SDK to start.
Switch ROI
Switch ROI, a Switch Trust capability, shows what every agent costs and recommends improvements, each with its estimated impact.See what every agent costs
Spending, caps, model pricing, and recommended improvements, in one place.
More
Connect your coding agent to Switch Trust
Bring Switch Trust data into Claude Code, Cursor, or VS Code. Fix an agent’s issues, cut its costs, and check that quality and safety held.
Scan and evaluate locally
Run Switch Trust CLI on your own machine. Free and source-available.
Connect your data sources
Choose the code repositories and cloud platforms Switch Trust scans for agents.
Browse your agent inventory
Every agent Switch Trust found, with its full record.
Route traffic with the Python SDK
Send your agents’ LLM traffic through Switch Trust.
Route traffic with the TypeScript SDK
Send your agents’ LLM traffic through Switch Trust from Node.js.
Look up a rule
Built-in discovery, MCP server, and agent runtime rules, with how to resolve each.
Query the API
Your AI inventory, findings, and rules over HTTP.