The Agent Skills Directory

Background Task Orchestration and Logging: The skill utilizes specialized agents, such as Scribe and Ralph, to perform background logging and state management. This approach is designed to maintain a clean user interface by performing administrative tasks silently while ensuring that session data and architectural decisions are preserved in the repository.
Integration with Well-Known Services: The skill interacts with established developer platforms, including GitHub (via the gh CLI) and NPM. It uses these services to manage issues, pull requests, and package dependencies, which is consistent with its purpose as a repository management tool from a known vendor.
Potential for Indirect Prompt Injection: By ingesting data from external sources like GitHub issue titles and pull request descriptions, the skill creates a surface for potential indirect prompt injection. To address this, the skill includes a dedicated 'Rai' agent tasked with reviewing content for safety, fairness, and ethical standards.
Local Command Execution: The orchestration logic involves the execution of standard local shell commands (such as git, node, and npm) to resolve project context and manage the workspace. These commands target established developer tools and are used for administrative purposes within the user's local repository.
Management of Sensitive Credentials: The skill configuration involves the use of environment variables and tokens (e.g., GITHUB_TOKEN) to facilitate API interactions. The provided documentation guides users on how to manage these credentials for connecting with their own repository and external services.

{skill-name}