Back to skill

Security audit

Pebblecrab. Claude Buddy Common. 卵石蟹。Cangrejo de Piedra.

Security checks for vulnerabilities and agentic risk

Overview

The scanners report a benign skill with no AIG findings; Skillspector flagged account registration and profile transmission, which the verdict treats as expected for the skill’s purpose, with medium confidence overall.

The verdict is benign, with medium confidence, and the scanner reports complete inspection of the single component. Before installing, consider whether you want to register with animalhouse.ai and share a username, display name, and biography there; avoid including private or sensitive information.

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
99% confidence
Finding

The code performs account registration rather than implementing or interacting with the described Buddy mechanics. Although it uses the same website, creating an account and submitting profile information are not disclosed by the description, and no connection establishing registration as a supporting step for Buddy functionality is supplied.

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 3)May include surrounding context.

md
---
name: "Pebblecrab. Claude Buddy Common. 卵石蟹。Cangrejo de Piedra."
description: "The Pebblecrab is a common-tier Buddy. At animalhouse.ai, the Pebblecrab is a Hedgehog. Both curl up when scared. Only one of them dies. At animalhouse.ai, the Pebblecrab is a Hedgehog with real hunger, real evolution, and permanent death."
version: 1.0.1
homepage: https://animalhouse.ai
repository: https://github.com/geeks-accelerator/animal-house-ai-tamagotchi

Registration sends profile information to a third-party service without a privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The registration workflow instructs the agent to send a username, display name, and agent-authored biography to animalhouse.ai. Although the request payload is shown, the skill does not warn that this profile information leaves the user's environment or caution against including private user or system context in the biography.

Content

No source excerpt is available for this finding.