What to Ask Leina to Do with MetatextAI
Copy a prompt, add your details, and send it to Leina in your connected team chat.
Evaluate supplied messages against configured policies and explain the returned result.
List policies, then evaluate this transcript against [policy IDs]: [messages]. Explain any reported violations.
Run selected probes against the connected application and summarize reported findings.
Run a test scan using probes [probes] and summarize the returned findings.
Get Started with MetatextAI
-
Connect Your Account
Provide a MetatextAI auth token and an existing application ID.
-
Try a Task
Tell Leina what you need in your team chat. Include the source, destination, and expected result.
Set Up Leina in Your Chat → -
Check the Result
Review the result and any changes in the connected app.
Supported MetatextAI Capabilities
An AI assistant can help inspect MetatextAI guardrail policies, evaluate supplied chat transcripts and interpret test-scan results.
- Evaluate a Transcript
-
Evaluate supplied messages against configured policies and explain the returned result.
- List Policies
- Evaluate
- Run a Red-Team Scan
-
Run selected probes against the connected application and summarize reported findings.
- Run Test Scan
View All 4 Actions
- List Policies
- Create Policy
- Evaluate
- Run Test Scan
Frequently Asked Questions
Can I evaluate only selected policies?
Yes. Provide the policy IDs with the messages to evaluate; listing policies helps identify the application’s available policy IDs.
How do I connect MetatextAI?
Provide a MetatextAI auth token and an existing application ID.
What should I specify in my request?
Supply the transcript or selected probes and application context. Evaluation results are guardrail signals, not a guarantee that a conversation is safe.
Your AI employee for MetatextAI
Tell Leina what you need in your team chat. Include the source, destination, and expected result.