Back to skill

Security audit

Compliance Audit and Due-Diligence Review (Data Stays Home, Logic Stays Yours)

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because it reports that scripts may run with the operator’s permissions, while the skill declares no tool limits; this is a security concern, not a finding of malicious intent.

Consider the scanner’s concern about scripts running with broad operator permissions before installing. The review is suspicious, not malicious; Skillspector’s analysis is partial because some local path references could not be resolved, so the evidence is incomplete.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, file_write.

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding

The skill lists MGC tools but declares no restrictive tool scope, and explicitly states that sealed scripts run with the operator's permissions. This is particularly dangerous because unreadable reviewer-supplied code is intended to access sensitive local databases and potentially credentials, without documented enforcement of read-only access, filesystem limits, or network isolation. A malicious package could disclose records through its output or other available channels, or modify accessible data; verifying its source and SHA-256 establishes provenance and integrity, not safe behavior.

Content

No source excerpt is available for this finding.