gitlab
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from GitLab, creating a surface for indirect prompt injection.\n
- Ingestion points: The scripts/gitlab.py wrapper and direct glab CLI commands retrieve data from issue descriptions, merge request bodies, and CI/CD logs.\n
- Boundary markers: The skill does not use protective delimiters or specific instructions to warn the agent about potentially malicious content in the retrieved data.\n
- Capability inventory: The skill enables high-privilege actions such as merging code, triggering CI/CD jobs, and managing repositories via the glab CLI.\n
- Sanitization: External data is formatted for markdown presentation but is not filtered to remove or escape potential instructions.\n- [COMMAND_EXECUTION]: The Python wrapper scripts/gitlab.py executes the glab CLI using the subprocess module to interact with GitLab resources.\n- [EXTERNAL_DOWNLOADS]: Utilizes the official GitLab CLI (glab) for all repository and project management operations.
Audit Metadata