ClawHub
Back to skill

Security audit

book-research

Security checks across malware telemetry and agentic risk

Overview

The skill is mainly a book research helper, but its included API script is broader than the stated Goodreads and Apple Books purpose.

Install only if you are comfortable with an agent having a generic Crawlora API helper under this skill. Prefer limiting use to the documented Goodreads and Apple Books endpoints and avoid passing private or unrelated data through the wrapper.

SkillSpector

By NVIDIA
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 (2)

Description-Behavior Mismatch

Medium
Confidence
96% confidence
Finding
The helper advertises and supports generic Crawlora API access rather than being constrained to the book-research capability described in the skill metadata. In an agent setting, this scope mismatch can let prompts or downstream callers use the skill to access unrelated endpoints and transmit arbitrary user data to external services under the skill's trusted identity.

Description-Behavior Mismatch

Medium
Confidence
99% confidence
Finding
The script accepts arbitrary HTTP methods, arbitrary API paths, arbitrary query parameters, and arbitrary JSON bodies, effectively exposing the full Crawlora API through this skill. Because the manifest frames the skill as book and audiobook research, this broad primitive materially increases abuse potential: a caller can repurpose the skill for unrelated data collection or external requests not anticipated by users or platform policy.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:22