The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown and natural language examples. It does not include scripts, executables, or hidden commands.
[SAFE]: No evidence of prompt injection or attempts to bypass safety filters were detected in the instructions.
[SAFE]: The instruction to check .agents/qa-project-context.md is a benign use of local context to inform the agent's behavior and does not constitute a data exposure risk.
[SAFE]: There are no external dependencies, downloads, or network requests performed by this skill.

qa-report-humanizer