ClawHub
Back to skill

Security audit

job-market-research

Security checks across malware telemetry and agentic risk

Overview

This skill is for job-market research, but its helper can use the Crawlora API key to call unrelated Crawlora endpoints, so it should be reviewed before installation.

Install only if you are comfortable giving the agent a Crawlora API key and relying on it to call only job-market endpoints. Prefer a constrained wrapper or careful user confirmation for each request, especially because the included helper can reach other Crawlora API paths and may consume account credits.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The helper advertises and enables broad Crawlora API usage with arbitrary paths and examples unrelated to job-market research, which exceeds the declared skill scope. In an agent setting, this scope mismatch can be abused to make authenticated requests to unrelated endpoints, increasing data exfiltration and capability-bypass risk.

Context-Inappropriate Capability

Medium
Confidence
97% confidence
Finding
The script concatenates a user-supplied path onto the API base and sends authenticated GET or POST requests to any Crawlora endpoint. Because the skill is supposed to research jobs and freelance gigs, unrestricted endpoint invocation creates an ambient-authority problem where the agent can be steered into using the API for unrelated or more sensitive operations.

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:23