doccoverage

Security checks across malware telemetry and agentic risk

Overview

DocCoverage is a local documentation scanner with disclosed license checks and optional git-hook setup, with no artifact evidence of hidden data transfer or destructive behavior.

Install if you want local documentation scanning. Be aware that Pro hook setup changes the current repository's lefthook configuration and can block commits; use it only in repos where you want that behavior and keep the license key in the documented environment variable or OpenClaw config.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (4)

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: The skill advertises itself as a documentation analyzer, but this code can modify repository configuration by creating or editing `lefthook.yml` and installing hooks. That is a real capability expansion beyond passive analysis, which increases trust and supply-chain risk because running the skill can persist behavior into the user's repo and affect future commits.

Context-Inappropriate Capability

Medium

Confidence: 89% confidence
Finding: Repository hook management is not necessary for a basic documentation coverage scanner, so its presence meaningfully broadens the attack surface. Even without obviously malicious logic, a skill that can alter git hook configuration can introduce persistent execution paths in developer workflows and undermine least-privilege expectations.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The trigger phrases are broad and overlap with common user requests about documentation, which can cause the skill to be invoked when the user did not clearly consent to running this specific tool. Because the skill can read local configuration, inspect repositories, and install or remove git hooks in some modes, accidental invocation increases the risk of unintended local actions or exposure of repository metadata.

Missing User Warnings

Low

Confidence: 72% confidence
Finding: The script accesses a local config file and an environment variable containing a license key without any explicit user-facing notice at the time of access. Although this appears intended for ordinary licensing, undisclosed secret/config access reduces transparency and can violate user expectations in a tool whose primary purpose is documentation analysis.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal