The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources that could contain malicious instructions designed to manipulate the agent.
Ingestion points: The skill reads PR review comments, inline comments, and issue comments using gh api and gh pr view (Step 3). It also retrieves CI logs using gh run view --log-failed (Step 4).
Boundary markers: The instructions lack explicit delimiters or warnings to the agent to ignore instructions embedded within the data it reads.
Capability inventory: The skill allows the agent to modify code (Step 6), commit changes, and push to a remote repository (Step 7).
Sanitization: No sanitization, filtering, or validation is performed on the external content before the agent interprets it to 'Address Valid Issues'.
[COMMAND_EXECUTION]: The skill relies on executing shell commands to interact with the local filesystem and remote GitHub repositories.
Evidence: Uses gh (GitHub CLI) for viewing PRs, checking CI status, fetching logs, and interacting with the GitHub API. It uses git for branch management, committing, and pushing changes.

iterate-pr