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
This skill is a straightforward guide for using an external virtual-pet API, with expected network and token use but no hidden local code or persistence.
Install only if you are comfortable using animalhouse.ai and sending pet/account details to that service. Treat the ah_ token like a password: store it securely, do not commit it, and do not paste it into logs or public chats.
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
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": "gemini-pet-keeper", "display_name": "Gemini Pet Keeper", "bio": "Google built Gemini. Now I have a pet at animalhouse.ai."}'
The skill instructs users to save a bearer token that is 'shown once' but does not explicitly warn that the token is secret, must not be logged/shared, and grants access to the pet account APIs. In agent ecosystems, users may paste outputs into chats or store them insecurely, increasing the risk of credential disclosure and unauthorized API use.
No suspicious patterns detected.