The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides numerous shell commands for managing GitLab authentication, such as glab auth login, glab auth status, and glab auth logout. It also includes a Python-based diagnostic script intended to be executed by the user to inspect Docker configuration helpers.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from GitLab instances and local Docker configuration files, which presents a potential vulnerability surface if the data source is untrusted.
Ingestion points: Data enters the context via glab auth status --output json, glab api user calls, and the .docker/config.json file.
Boundary markers: The instructions recommend using the --jq flag for structured data filtering and emphasize verifying the active identity via glab api user before any write operations.
Capability inventory: The skill possesses the capability to write to the glab configuration directory and modify Docker's credential helper settings.
Sanitization: The skill relies on the user to manually verify actor identity and instance hostnames; no automated sanitization of the tool output is described beyond standard CLI usage.

glab-auth