The Agent Skills Directory

[COMMAND_EXECUTION]: Executes standard development toolchains including npm, npx, and pytest for mandatory preflight checks such as building, linting, and testing.- [COMMAND_EXECUTION]: Interacts with repositories using git and the GitHub CLI (gh) to manage branches, commits, and pull requests.- [COMMAND_EXECUTION]: Employs shell HEREDOCs to safely handle commit messages, which prevents potential command injection vulnerabilities when processing generated text.- [DATA_EXFILTRATION]: Implements explicit protection against the exposure of secrets by specifically prohibiting the staging of .env files, credentials, and API keys.- [PROMPT_INJECTION]: Evaluates file diffs to generate commit messages and PR summaries. While this represents an indirect prompt injection surface, it is considered a low-risk, standard functionality of the coding assistant, and the skill uses boundary markers for safe shell integration.- [SAFE]: No malicious patterns, obfuscation, or unauthorized network operations were detected. The skill's behavior is consistent with its stated purpose as a deployment tool.

Ship