The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources including bug descriptions, game design documents, and sprint plans. While this represents an indirect prompt injection surface, the risk is minimized as the skill only outputs to a markdown manifest and does not have the capability to execute code or access sensitive system configuration files. Evidence: The skill reads from design/gdd/systems-index.md and project backlog tasks to identify testing gaps.- [SAFE]: All identified file operations and tool usages are consistent with the skill's stated purpose of managing a regression test suite. The skill uses Glob, Grep, Read, and Write tools exclusively on project-specific paths like tests/ and design/. No network activity or external downloads are performed.

regression-suite