ClawHub
Back to skill

Security audit

Universal Release

Security checks across malware telemetry and agentic risk

Overview

This is a release helper, but it can turn vague push requests into repository edits, commits, tags, and remote pushes beyond what a user may have intended.

Install only if you want an agent to manage full release preparation, not simple git pushes. Prefer dry-run first, review all file edits and commits before proceeding, and do not use this skill as the default handler for vague "push" requests.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill is presented as a release workflow, but it instructs the agent to perform broader repository maintenance by scanning READMEs, modifying documentation, and creating multiple per-module commits before the release. That scope expansion increases the chance that a simple user request like 'release' or 'push' causes unintended codebase changes and extra commits beyond the user's intent.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger phrases are broad enough to match ordinary requests such as 'push' or 'new version,' which can activate a high-impact workflow that edits files, creates commits, tags releases, and may push to remote. In the context of an agent skill, overbroad activation is dangerous because it can cause destructive or unexpected repository operations from ambiguous user input.

Vague Triggers

Medium
Confidence
96% confidence
Finding
The 'When to Use' guidance explicitly routes vague requests like 'push to remote' or 'just push' into the full release workflow, even when that workflow performs additional analysis, file modification, commits, tagging, and optional pushing. This mismatch between user intent and skill behavior makes accidental release actions substantially more likely, especially because git push is a sensitive operation with irreversible external effects.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.