The Agent Skills Directory

Indirect Prompt Injection: The skill is designed to ingest and act upon GitHub issues, which are external data sources. Content within these issues could potentially contain instructions aimed at influencing the agent's logic during the triage, research, or writing phases. This is a common consideration for agents that process untrusted text.
Command Execution: The skill utilizes the GitHub CLI (gh) and git to manage issues and repository state. It interpolates the issue identifier directly into command arguments. While this is necessary for the skill's functionality, it establishes a dependency on the integrity of the input provided to the command-line tools.
Automated Repository Interaction: The skill has the capability to modify code, create branches, and open pull requests. These actions are performed through a series of specialized sub-agents (triage, research, write, review), representing a complex autonomous workflow.

fix-issue