The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external GitLab resources which may contain malicious instructions.
Ingestion points: SKILL.md identifies multiple ingestion points where untrusted data enters the agent context, including merge requests (gitlab-mr), issues (gitlab-issue), repository files (gitlab-file), and discussions (gitlab-discussion).
Boundary markers: The instructions do not define boundary markers or explicit prompts to ignore instructions embedded within the GitLab data being retrieved.
Capability inventory: The skill utilizes the Bash, Read, Glob, and Grep tools to interact with the file system and execute shell commands.
Sanitization: There is no evidence of sanitization, validation, or escaping of content retrieved from GitLab before it is processed by the agent.
[COMMAND_EXECUTION]: The skill relies on the Bash tool to execute glab CLI commands for automation tasks.
Usage: Explicitly instructs the agent to use the Bash tool for operations like glab auth login, glab mr create, and glab repo clone as defined in SKILL.md.

gitlab-assistant