The Agent Skills Directory

[DATA_EXPOSURE]: The skill performs codebase exploration to understand architecture, module boundaries, and integration points. This involves reading repository files and test coverage, which is a standard and necessary function for creating a refactoring plan.
[COMMAND_EXECUTION]: The skill utilizes external command-line tools, specifically gh-cli for GitHub and az-devops-cli for Azure DevOps, to create issues and work items. These operations are triggered at the final step of the workflow to persist the refactoring plan to the user's chosen destination.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it reads untrusted data from the repository during exploration. * Ingestion points: Source code and documentation files read by the subagent in Step 2 of the workflow. * Boundary markers: None explicitly defined in the skill instructions to separate repository content from the agent's internal logic. * Capability inventory: Reading local files, writing local files to ./plans/, creating GitHub issues (gh issue create), and creating Azure DevOps work items (az boards work-item create). * Sanitization: No explicit sanitization or filtering of the content read from the codebase is mentioned before it is processed by the agent.

refactoring-plan