Security audit

context-map

Security checks across malware telemetry and agentic risk

Overview

This skill coherently helps an agent scan a project for context, but users should know it can inspect the repository and may create .codesight files.

Install this only if you are comfortable with an agent scanning the current project and possibly writing .codesight/ context files. Use --no-wiki when you want a read-only scan, and review or ignore .codesight/ in version control if it is generated.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger list includes broad terms such as "context," "scanning," and "exploration," which are likely to match many ordinary user requests and may cause this skill to activate unexpectedly. Because the skill directs the agent to run project-wide scanning commands, overbroad activation increases the chance of unrequested command execution and unnecessary codebase enumeration.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill documents that the scanner generates wiki articles under `.codesight/`, but it does not clearly warn users in the main description or procedure that invoking the skill will write files to the working tree. Undisclosed filesystem modification is security-relevant because it can surprise users, alter repository state, pollute commits, or create trust issues in environments expecting read-only exploration.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.