The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local shell commands including git, gh (GitHub CLI), npm, and curl to manage the software development lifecycle, sync branches, and verify deployment status.
[COMMAND_EXECUTION]: Implements identity verification steps using git config and git log to ensure the agent is operating as the authorized user and to prevent local identity overrides.
[COMMAND_EXECUTION]: Automated monitoring of production health and deployment status is performed via gh api calls and curl requests to the production endpoint.
[PROMPT_INJECTION]: The instructions contain multiple internal safety constraints (e.g., 'Never push on your own', 'Only run this after the user explicitly told you to push') which act as guardrails against autonomous unauthorized actions.
[SAFE]: Ingestion of data from external APIs and local logs is handled using structured parsing tools like jq, which reduces the potential surface for indirect prompt injection from untrusted source data.

prod-push