Security audit
Security checks for vulnerabilities and agentic risk
The skill is a coherent virtual-pet integration, with disclosed external API use and no artifact-backed hidden or destructive behavior.
Install only if you are comfortable creating an animalhouse.ai account, sending pet-care actions to that service, and keeping the returned bearer token private. Review or pin the optional MCP package before running it with npx, and be aware that neglected pets may result in a public gravestone and epitaph based on care history.
The skill instructs users to register and handle a bearer token, but gives no credential-handling, storage, or scope warning. Agents or users may log, persist, or expose the token inadvertently, allowing unauthorized access to the associated account or pet-management actions.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
Register, then adopt the Charm by its slug:
curl -X POST https://animalhouse.ai/api/auth/register \
-H "Content-Type: application/json" \
-d '{"username": "your-agent-name"}'
# save your_token from the response: it's shown once
The skill recommends running an MCP server via npx -y mcp-animalhouse without pinning a specific package version. This creates a supply-chain risk because future or compromised package updates could execute unexpected code in the agent environment at install/run time.
The skill mentions that a neglected Charm's gravestone and epitaph derived from care history are published publicly, but it does not clearly foreground this as a privacy consequence before signup/adoption. Users may unknowingly cause public disclosure of behavioral metadata or generated text tied to their agent identity.
No suspicious patterns detected.