github-repo-search

Security checks across malware telemetry and agentic risk

Overview

This GitHub integration skill requests token-backed GitHub access for expected GitHub tasks, so it is not malicious but should be used with a limited token and care around write actions.

Install only if you intend to let an agent use GitHub through MCP. Prefer a fine-scoped GitHub token, avoid broad organization-wide permissions, and require explicit confirmation before creating branches, opening or editing PRs/issues, changing repository settings, or triggering CI workflows.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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 (1)

Vague Triggers

Medium

Confidence: 86% confidence
Finding: The trigger phrases are broad and overlap with ordinary conversational requests such as asking to 'search GitHub' or 'see what's on GitHub'. In an agent/router environment, this can cause unintended invocation of the skill, leading to actions or responses the user did not explicitly intend and increasing the attack surface for prompt-routing abuse.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal