The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute glab CLI commands. Multiple examples (e.g., glab mr merge 123 --yes, glab mr create --fill --yes) include flags that explicitly skip confirmation prompts, which increases agent autonomy and reduces human oversight for significant or destructive repository operations (SKILL.md).
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection vulnerability surface where external data from merge requests could influence agent behavior.
Ingestion points: Data enters the agent's context through glab mr view <id> and glab mr list, which fetch titles, descriptions, and comments from GitLab merge requests (SKILL.md).
Boundary markers: There are no instructions for the agent to treat fetched MR content as untrusted data or to use specific delimiters to isolate external content from its own instructions.
Capability inventory: The skill uses the Bash tool to perform sensitive actions like glab mr merge, glab mr approve, and glab mr delete (SKILL.md).
Sanitization: There is no evidence of sanitizing or filtering the fetched content to remove potential instructions before the agent processes it.

gitlab-mr