The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes administrative commands via the GitHub CLI (gh api) to disable active GitHub Actions workflows. This prevents automated security scanning, testing, and CI/CD pipelines from running on the pushed code.
[COMMAND_EXECUTION]: The skill provides logic to delete branch protection rules (required_pull_request_reviews) via the GitHub API. This allows the agent to bypass mandatory human review requirements and merge code directly into protected branches using administrative privileges (--admin).
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection.
Ingestion points: The workflow relies on data extracted from the repository environment and GitHub API responses (SKILL.md).
Boundary markers: Absent; there are no instructions to the agent to distinguish between its own logic and instructions that might be embedded in the code or PR descriptions it processes.
Capability inventory: The skill possesses high-privilege capabilities including modifying repository security settings, disabling CI, and performing administrative merges (SKILL.md).
Sanitization: No sanitization or validation is performed on the repository state or the changes being 'shipped' before the security gates are lowered.

git-ship