The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data originating from GitHub issues and pull request comments.
Ingestion points: Fetches issue data (gh issue view) and review comments (gh api) from the homeassistant-ai/ha-mcp repository.
Boundary markers: The instructions lack explicit delimiters or warnings to treat the content of issues or comments as potentially untrustworthy text.
Capability inventory: The skill possesses the ability to modify the local filesystem (Read, Edit, Write tools) and perform network operations via the GitHub CLI and Git, including pushing code to remote branches.
Sanitization: There is no evidence of sanitization or filtering of the content retrieved from GitHub before the agent processes it.
[COMMAND_EXECUTION]: The skill utilizes shell commands for managing the development lifecycle.
Evidence: It uses git for worktree and branch management, and gh for interacting with the GitHub API and managing pull requests. All operations are targeted at the vendor's official repository.
[DYNAMIC_EXECUTION]: The skill executes the repository's test suite during the implementation phase.
Evidence: The instruction uv run pytest src/e2e/ triggers the execution of Python code within the generated worktree. This is a primary function of the skill to ensure implementation quality.

issue-to-pr-resolver