Security audit

git-weekly

Security checks across malware telemetry and agentic risk

Overview

This is a read-only Git weekly-report skill, but it can read recent code diffs when triggered.

Install only if you are comfortable with the agent reading recent Git history and full diffs from the current repository. Use an explicit trigger such as “git-weekly”, avoid broad summary phrases when you do not want repository analysis, and review generated reports for sensitive code or internal details before sharing.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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 phrases include very generic inputs like “生成总结”, which can easily appear in ordinary conversation and unintentionally activate the skill. Because this skill automatically runs `git log --patch` over the last 7 days, accidental invocation could expose commit messages and code diffs without the user clearly intending to share repository history.

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: The skill instructs automatic execution of `git log --since="7 days ago" --author="$(git config user.name)" --reverse --patch` but does not warn the user that recent git history and full patches will be read and analyzed. This is dangerous because patches may contain sensitive code, secrets, internal URLs, incident details, or proprietary implementation information, and the access happens implicitly rather than with informed consent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.