gitlab-create-mr
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes local repository data such as commit messages, diffs, and branch names to automatically generate merge request content. This content is then used as input for shell commands like glab mr create. Although this introduces a surface for indirect prompt injection via malicious repository data, the skill includes explicit instructions for the agent to sanitize the output. Evidence: 1. Ingestion points: SKILL.md workflows utilize git log and git diff for context gathering. 2. Boundary markers: SKILL.md and references/mr-body-conventions.md contain specific rules to exclude sensitive data. 3. Capability inventory: The skill utilizes git and glab for repository and platform interactions. 4. Sanitization: The agent is directed to remove local debugging info, secrets, and machine-specific paths.
- [COMMAND_EXECUTION]: The skill's primary functionality is delivered through the execution of local shell commands, specifically git and glab. These commands are used for status checks, log analysis, and merge request management as documented in references/verified-commands.md.
Audit Metadata