The Agent Skills Directory

[SAFE]: The skill is designed for static analysis of test metrics and logs. No malicious patterns such as remote code execution, credential harvesting, or unauthorized network operations were detected.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface by processing external test logs and reports supplied by the user. However, the risk is minimal because the instructions strictly forbid the agent from executing code, editing files, or taking autonomous actions based on the input. It also mandates a structured output that separates factual evidence from inferences, which acts as a logical boundary against instruction overrides embedded in processed data.

test-suite-health-analysis