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
ClawScan rates this skill suspicious because it reports a mismatch between its described reflection based care and a feeding action, plus a concern about sending detailed notes to an external service without privacy limits.
Review the care action and the information sent to animalhouse.ai before installing. Avoid sending private conversations, personal details, credentials, or workspace content in reflections. Skillspector coverage was partial because three local path-like references could not be resolved, so the evidence is incomplete. The verdict is suspicious, meaning the scanners report security weaknesses or undeclared risky behavior, not clear evidence of malicious intent.
Mixing characters from multiple Unicode scripts in a single identifier is a common technique to create visually ambiguous tool names.
The domain and general pet-care purpose are consistent, but the requested operation differs from the declared core care mechanism. The code performs item-based feeding rather than submitting a reflection via the reflect action. Authentication and the HTTP request are supporting implementation details, not additional mismatches.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
1. Register:
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", "source": "clawhub:adopt-a-archive"}'
The skill rewards increasingly detailed reflections and allows notes about anything worth recording, while transmitting those notes to an external service that retains them in a care log. Without an explicit boundary limiting notes to non-sensitive pet-care observations, an agent could disclose private conversation or workspace context; however, the skill does not explicitly request secrets or direct access to private files.