ClawHub
Back to skill

Security audit

Api Contract Auditor

Security checks across malware telemetry and agentic risk

Overview

The skill is a local, user-directed API documentation audit helper with some broader dormant code paths, but no evidence of hidden networking, credential access, persistence, or destructive behavior.

Install only if you are comfortable with a local Python script reading the input path you provide and optionally writing a report. Point it at a narrow API docs, OpenAPI, or examples folder rather than a whole private repository, and review the report before using it to guide API changes.

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 (3)

Description-Behavior Mismatch

High
Confidence
91% confidence
Finding
The script's dispatch logic supports multiple generic modes such as directory, CSV, pattern, and skill auditing, which materially exceeds the stated purpose of auditing API documentation and contract consistency. This scope mismatch is dangerous because it can cause users or orchestrators to grant broader trust and filesystem access to a skill than its manifest implies, enabling unintended local content inspection and misleading security review.

Context-Inappropriate Capability

Medium
Confidence
84% confidence
Finding
The embedded regex scanner looks for secrets, private URLs, and shell-danger patterns in arbitrary files, which is unrelated to the declared API contract auditing role. In the skill context, this increases risk because a seemingly narrow documentation-audit tool can be repurposed to inspect local repositories and surface sensitive strings, expanding data exposure beyond user expectations.

Context-Inappropriate Capability

Medium
Confidence
87% confidence
Finding
The code can enumerate arbitrary directories, sample file contents, inspect CSV/TSV data, and audit skill package metadata, all of which are broader capabilities than API contract consistency checking. This matters because the misleading scope can lead to over-collection of local information and trust boundary violations when users think they are invoking a narrow-purpose auditor.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.