OpenProject by altf1be

• Outcome: Pass
• Latest audit: May 29, 2026, 8:17 PM
• Version:
  2.0.0

Overview

This is a broad but clearly documented OpenProject management skill that uses an API token to run user-invoked OpenProject API actions.

Install only if you want an agent to manage OpenProject on your behalf. Use a dedicated least-privilege API token, review create/update/delete/custom-action/OAuth commands before running them, be careful with --all notification changes, and only upload files you intend to send to the configured OpenProject instance.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Intent-Code Divergence

Medium

• Confidence: 97% confidence
• Finding: The comment claims there is no arbitrary file access, but attachment-upload commands can read any local path supplied by the caller via --file. The safePath helper only blocks '..' segments and still allows absolute paths or arbitrary files within accessible directories, so a prompt-injected or mistaken caller could exfiltrate sensitive local files to the remote OpenProject instance.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: Bulk notification mark-read/unread operations can affect all matching notifications without any confirmation prompt or dry-run output. In an agent context, a malicious or erroneous instruction could silently alter a user's notification state at scale, causing missed alerts or workflow disruption.

VirusTotal

66/66 vendors flagged this skill as clean.