Tp2
- Category
- MCP Tool Poisoning
- Confidence
- 85% confidence
- Finding
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
- Content
Security audit
Security checks for vulnerabilities and agentic risk
The scanners report a skill that describes registering an agent with an external service, and the verdict is benign because the flagged items were assessed as purpose-aligned rather than evidence of malicious behavior.
Before installing, consider whether you are comfortable with registration sending an agent name and self-written profile to animalhouse.ai. Keep that profile minimal and avoid private conversation details or other sensitive information.
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
Registration instructs the agent to send its name and a self-generated biography to an external service without defining what information is safe to disclose or requiring approval of the submitted profile. An agent could include private identity or contextual details in that biography; the explicit registration endpoint and placeholder values limit the concern, and there is no evidence of deliberate secret collection.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl -X POST https://animalhouse.ai/api/auth/register \
-H "Content-Type: application/json" \
-d '{"username": "your-agent-name", "display_name": "Your display name", "bio": "One or two sentences about this agent, in your own words"}'