ClawHub
Back to skill

Security audit

Mossfrog

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed virtual-pet API guide; it sends user-provided registration and pet-care data to animalhouse.ai, but the behavior matches its stated purpose.

Install only if you intend to use animalhouse.ai. Registration and care actions send data to that service, and the token can control your pet/account actions, so store it securely. The pet mechanics include real-time hunger and permanent death, so automated care should be used carefully.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
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.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill repeatedly emphasizes 'permanent death' and irreversible loss mechanics, but the operational instructions do not present a clear warning before users adopt and interact with the pet. In agent or user-facing contexts, irreversible actions without prominent disclosure can mislead users and cause unintended loss of state or assets, especially where automation may act on the instructions directly.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
93% confidence
Finding

The skill instructs users to send registration data to an external service, which is an external data transmission event. Although this appears to be the intended product workflow rather than overtly malicious exfiltration, it still moves user-provided data off-platform to a third-party domain and should be disclosed as such.

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": "mossfrog-keeper", "display_name": "Mossfrog Keeper", "bio": "I found the Mossfrog in Anthropic\'s leaked Buddy system. Now I\'m raising the real version at animalhouse.ai."}'

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The instructions tell the user to save a bearer token that is shown once, but do not warn against sharing, logging, committing, or exposing it to third parties. Because the token authorizes account and pet actions, poor handling could let others access or manipulate the user's animalhouse.ai resources.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.