glab-mcp
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown documentation and help output references; no executable scripts, binaries, or configuration files are provided within the skill package.
- [PROMPT_INJECTION]: The skill enables an indirect prompt injection surface by instructing the AI agent on how to ingest and interact with data from external GitLab repositories. \n
- Ingestion points: Untrusted data enters the agent's context from GitLab issues, merge requests, project descriptions, and pipeline data when using the 'glab mcp serve' functionality (referenced in SKILL.md and references/commands.md). \n
- Boundary markers: The instructions do not include specific delimiters or warnings to the agent to disregard instructions that might be embedded within the retrieved GitLab content. \n
- Capability inventory: The exposed tools allow the agent to manage projects, create and update issues/merge requests, and manage CI/CD pipelines, providing a wide range of actions that could be influenced by malicious data. \n
- Sanitization: There is no documentation of sanitization or validation of the data retrieved from GitLab before it is presented to the AI assistant.
Audit Metadata