AGENT LAB: SKILLS

gitlab-automation

Warn

Audited by Snyk on Feb 15, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.80). This skill ingests and displays user-generated GitLab content via the GitLab toolkit (e.g., GITLAB_LIST_PROJECT_ISSUES, GITLAB_GET_PROJECT_MERGE_REQUESTS, GITLAB_LIST_REPOSITORY_COMMITS, GITLAB_GET_USERS) through the Rube MCP connection, so it will read arbitrary third‑party issue/MR/commit/comment text that could contain untrusted instructions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.90). The skill requires contacting the MCP server at https://rube.app/mcp at runtime (via RUBE_SEARCH_TOOLS / RUBE_MANAGE_CONNECTIONS) to fetch tool schemas that directly determine the agent's prompts/instructions, making it a runtime external dependency that can control behavior.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 15, 2026, 08:20 PM