Back to skill

Security audit

Anthropic Pet. Buddy Companion for AI Agents. 人类公司宠物。Mascota Anthropic.

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: it reports an external account registration that sends an agent name, display name, and biography, while the verdict treats the flagged behavior as purpose-aligned.

Before installing, consider whether you want to register with animalhouse.ai and send it an agent name, display name, and biography. The scan reports complete file coverage, but Skillspector also returned a non-zero exit status with serializer warnings; the overall verdict has medium confidence.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Tp2

High
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

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The destination matches the advertised website, but the description does not disclose account creation or profile submission. The supplied code performs registration rather than demonstrating or interacting with the described pet features. This is an undeclared, state-changing capability, not merely an implementation detail of the advertised features.

Content

No source excerpt is available for this finding.

Registration transmits agent identity and biographical information to a third-party service without a privacy warning.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The registration workflow instructs the agent to submit a username, display name, and self-written biography to animalhouse.ai. Unlike the endpoint catalog, this is an actionable step that sends information externally, but the skill does not warn the user about that disclosure or advise against including private contextual information.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 50)May include surrounding context.

Register

bash
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"}'