The Agent Skills Directory

[COMMAND_EXECUTION]: Employs standard git and gh (GitHub CLI) commands to perform repository inspections, branch management, and API interactions. These operations are necessary for the skill's functionality and are confined to the local repository and GitHub environment.- [DATA_EXFILTRATION]: All network operations are directed toward official GitHub endpoints via the gh CLI for repository metadata and pull request creation. No unauthorized data transmission to third-party or suspicious domains was identified.- [CREDENTIALS_UNSAFE]: The instructions include a specific safety directive to 'Exclude secrets and unrelated local changes' during the staging process, mitigating the risk of accidental credential exposure in commits.- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local repository (file diffs and PR templates) to generate pull request content. This ingestion is limited to the skill's primary purpose and is managed through structured API calls rather than direct execution of the external content.

create-pr