The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted data from the codebase being triaged and uses that data to generate a GitHub issue.
Ingestion points: The skill reads source files, git logs, and existing tests from the user's codebase in the 'Explore and diagnose' phase (SKILL.md).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat code comments or documentation as data rather than instructions during exploration.
Capability inventory: The skill has the capability to read local files via the Agent tool and write content to external platforms using the gh issue create command.
Sanitization: No sanitization or escaping of the content retrieved from the codebase is performed before it is interpolated into the GitHub issue template.

triage-issue