ClawHub
Back to skill

Security audit

consensus-persona-engine

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be a small, disclosed Node.js reputation-update engine with no network access, credential use, file writes, or persistence.

Review the reputation rules before relying on the output for real governance decisions, but the artifact behavior is clearly scoped to local JSON input and deterministic in-memory reputation updates.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding
The declared description is for a specific reputation-processing skill, but the supplied code chunk only shows a launcher script. Its observable behavior is to access the filesystem, load JSON input, invoke an external handler, and emit output. Because the core reputation logic is not present in this chunk, the code shown does not substantiate the declared primary purpose. It also introduces an undeclared resource access capability (local file reads) and a CLI trigger surface. While file input may be a support detail in a full application, based on this chunk alone the actual behavior does not accurately match the specialized declared description.

Static analysis

No suspicious patterns detected.