gitlab-automation

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow uses Composio’s GitLab toolkit calls (e.g., GITLAB_LIST_PROJECT_ISSUES, GITLAB_GET_PROJECT_MERGE_REQUESTS, GITLAB_LIST_PROJECT_PIPELINES) which return GitLab issue/MR/pipeline text authored by other users (outsiders) into the agent context as readable fields like titles/descriptions/comments, enabling indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). This skill requires adding and using the external MCP endpoint https://rube.app/mcp at runtime (via RUBE_SEARCH_TOOLS / RUBE_MANAGE_CONNECTIONS) to fetch tool schemas and control the agent's available prompts/instructions, so it is a runtime external dependency that directly controls agent behavior.