The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several git and GitHub CLI commands to manage repository state and PR lifecycle.
Evidence: Execution of git push, gh pr create, gh stack link, gh project item-list, and various gh api calls to interact with the GitHub API.
Purpose: These commands are essential for the skill's stated purpose of automating the PR process.
[INDIRECT_PROMPT_INJECTION]: The skill processes external, potentially untrusted data from the repository to generate PR metadata and configure project updates.
Ingestion points: Data is read from commit messages via git log (used for PR titles and bodies), existing PR details via gh pr list, and local repository files such as docs/agents/issue-tracker.md for project status configuration.
Boundary markers: There are no explicit boundary markers or instructions to the model to ignore potential commands embedded in the commit messages or configuration files.
Capability inventory: The skill has the ability to create pull requests, update project items, and perform API calls based on this interpolated data.
Sanitization: No specific sanitization or validation logic is defined for the content extracted from commit messages or local configuration files before it is used in tool arguments.

open-pr