The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands including git, awk, sed, and the glab (GitLab CLI) to automate branch management, conflict detection, and Merge Request creation. These actions are performed to fulfill the skill's primary purpose of synchronizing branches.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it reads untrusted data from the local and remote git repository to generate content for commits and Merge Requests.
Ingestion points: Data enters the agent context through git log, git ls-remote, git status, and git diff commands executed in SKILL.md.
Boundary markers: The instructions mandate a manual confirmation step ("Step 6 — Push gate") where the agent must wait for user approval before pushing code to the remote repository.
Capability inventory: The skill has the capability to write to the repository (git commit, git push) and interact with the GitLab API (glab mr create).
Sanitization: There is no explicit logic described to sanitize or escape branch names, commit messages, or file diffs before they are interpolated into shell command arguments for git commit or glab mr create.

gitlab-sync