The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute glab and git commands for repository interaction. Evidence includes glab mr view, glab mr diff, glab ci trace, and git push as defined in SKILL.md.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it reads and processes untrusted data from external GitLab repositories.
Ingestion points: Data is fetched from GitLab Merge Request descriptions, comments, and CI logs using glab mr diff, glab mr note list, and glab ci trace.
Boundary markers: While the skill uses markdown headers and delimiters to organize MR data in sub-agent prompts, it lacks explicit safety instructions for the AI to ignore potential commands embedded within the fetched content.
Capability inventory: The skill is capable of executing shell commands (glab, git) and spawning sub-agents with the context of the ingested data.
Sanitization: The skill does not perform explicit sanitization or filtering of content retrieved from GitLab before passing it to the agent or its sub-agents.

gitlab-claude