Back to skill

Security audit

Puddlefish. Claude Buddy Common. 水坑鱼。Pez Charco.

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign with medium confidence: scanners report a skill presented as a virtual duck experience that also includes account registration and an adoption request, with concerns about disclosure and data sharing.

Before installing, consider whether you are comfortable with the reported registration and adoption requests, and with sending a username, display name, and biography to animalhouse.ai. The verdict is benign, not risk-free. Skillspector's run includes an exit-status error despite reporting complete coverage, so treat its evidence with some caution.

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 (5)

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
99% confidence
Finding

The description presents a virtual pet but does not disclose account registration. The supplied code exclusively attempts registration and contains no pet interaction or lifecycle behavior. Although the destination domain matches the description, registration is a materially distinct, undeclared operation rather than an implementation detail of any described action.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The destination and creature details align with the description, but the description only characterizes the creature; it does not disclose that the skill initiates an adoption through an authenticated API. The code uses a placeholder token, so successful execution would require a valid token. It does not itself implement hunger, evolution, or permanent death.

Content

No source excerpt is available for this finding.

The registration workflow sends agent-identifying information to a third-party service without a privacy warning.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The registration instructions submit a username, display name, and agent biography to animalhouse.ai. This is an actionable workflow rather than an endpoint listing, but it does not warn that this information will leave the user's environment and be shared with an external service.

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 48)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"}'