The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute GitHub CLI (gh) commands to list and view issues, view pull request details, and perform merges. This includes the use of the --admin flag to bypass branch protections, which is a powerful command within the GitHub ecosystem.
[INDIRECT_PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection due to its processing of untrusted external content.
Ingestion points: In references/github-pr-operations.md, the skill uses gh issue view and gh pr view to fetch titles, bodies, and comments from external GitHub issues and pull requests.
Boundary markers: No boundary markers or "ignore embedded instructions" warnings are specified when the agent is instructed to read and incorporate this external data.
Capability inventory: The skill has the capability to write back to the environment by drafting PR bodies and merging pull requests.
Sanitization: There are no instructions provided for sanitizing, escaping, or validating the content retrieved from GitHub before it is processed into new PR descriptions.

pull-request