The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the glab CLI tool to perform repository inspection (glab repo view), merge request viewing (glab mr view), and raw API interactions (glab api). These commands are standard for GitLab management tasks and leverage the user's existing local environment.
[EXTERNAL_DOWNLOADS]: Installation instructions in the documentation reference fetching the skill from the author's own domain (https://skills.flc.io) and GitHub repository (github.com/flc1125/skills). These are recognized as legitimate vendor resources for this skill.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external GitLab API responses to resolve merge request versions and diff state, which could theoretically contain malicious content intended to influence agent behavior.
Ingestion points: Merge request versions and diff data fetched from the GitLab API via glab api calls.
Boundary markers: None explicitly defined in the prompt instructions for isolating API output.
Capability inventory: The skill has capabilities to read repository metadata and write comments to merge request discussions using glab.
Sanitization: The skill implements strong validation logic, requiring exact matching of SHAs, file paths, and line numbers against the current diff state before proceeding with any write operations.

gitlab-mr-inline-comment