The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the git and glab (GitLab CLI) tools to manage the development lifecycle. It executes commands to resolve repository information, create branches, update issue descriptions, and create merge requests. These actions are performed within the scope of the skill's intended functionality and include user-in-the-loop checkpoints before merging or committing changes.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external GitLab issues via glab issue view to determine implementation steps and update project status. This external content could potentially contain malicious instructions intended to influence the agent's behavior.
Ingestion points: GitLab issue descriptions and checklists are fetched from a remote server (gitlab.com or a self-hosted instance) in SKILL.md.
Boundary markers: The skill instructs the agent to focus strictly on a specific 'Phases' checklist structure to scope its implementation actions, which acts as a logical boundary.
Capability inventory: The agent has the capability to write and modify files in the repository, execute shell commands via Git/GitLab CLI, and interact with the user.
Sanitization: There is no evidence of explicit sanitization or filtering of the content fetched from GitLab issues before the agent processes the instructions contained within them.

codely-plan-phase-implement-gitlab