ClawHub
Back to skill

Security audit

base64-toolkit

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward Base64 utility, with file read/write helpers that fit its purpose but require careful path selection.

Install only if you need Base64 utilities. When using file encode/decode helpers, provide explicit trusted paths and avoid pointing decodeToFile at existing important files unless you intend to overwrite them.

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 (3)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill metadata advertises only base64 encoding/decoding/validation, but the implementation also exposes arbitrary filesystem read and write helpers via encodeFile() and decodeToFile(). In an agent/tooling context, this expands capability beyond user expectations and can enable reading sensitive local files or overwriting files if an upstream caller passes attacker-controlled paths.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

decodeToFile() writes attacker-controlled decoded content directly to an arbitrary path with fs.writeFileSync() and provides no warning, confirmation, or safety checks. In agent usage, this can silently overwrite existing files or drop untrusted content onto disk, increasing the risk of destructive writes or planting files in sensitive locations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

This markdown file advertises "File encoding/decoding" as a feature, which implies operations affecting user files, but it provides no warning or disclosure about reading from or writing to files. For markdown files, safety-relevant behaviors that could affect user data should be accompanied by a brief warning so users understand the impact.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.