The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown documentation, instructions, and examples. It does not contain any executable scripts, binaries, or configuration files that could perform actions on a system.
[SAFE]: No network operations, sensitive file path accesses, or hardcoded credentials were found in the provided content.
[INDIRECT_PROMPT_INJECTION]: The skill instructs an agent to process and analyze external content (task deliverables and test results). While this creates a theoretical surface for indirect prompt injection from untrusted data, the skill itself provides no dangerous capabilities for an injection to exploit, and its primary purpose is verification, which acts as a defense mechanism.

completion-verifier