ClawHub
Back to skill

Security audit

Semantic Code Search

Security checks across malware telemetry and agentic risk

Overview

This is a local code-search helper that reads a chosen codebase and may save a local index containing copied source snippets.

Install only if you are comfortable with it indexing the repositories you point it at. Do not run it on codebases containing secrets you do not want copied into a local JSON index, keep .code_index.json out of commits and shared folders, and verify the correct CLI entrypoint because the docs and included script name differ.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
83% confidence
Finding
The skill describes capabilities that read from a codebase path and build an index from repository contents, and it also implies writing index artifacts, but it declares no permissions. That mismatch can cause the platform or reviewers to underestimate the skill's data access and modification capabilities, which is risky because code search naturally touches potentially sensitive source files and may persist derived data.

Vague Triggers

Medium
Confidence
75% confidence
Finding
The description uses broad activation wording like 'Use when searching code by meaning rather than text' and 'navigating unfamiliar codebases,' which could cause the agent to invoke this skill for many generic coding tasks. Over-broad routing increases exposure of repository contents to the skill and can bypass more specialized or least-privileged tools, especially in environments with sensitive codebases.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.