The Agent Skills Directory

[SAFE]: The skill's primary function is the analysis of engineering data to produce quality metrics. All identified scripts and configurations are focused on this purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources such as commit histories and incident reports which represents a potential attack surface for indirect prompt injection.
Ingestion points: Metric data inputs defined in evals/quality-metrics.yaml and processed by agents.
Boundary markers: None explicitly defined for untrusted data interpolation.
Capability inventory: Coordination with multiple agents including qe-quality-gate which can block code deployments based on metric results.
Sanitization: Input data is validated against the schema defined in schemas/output.json but no specific string sanitization for natural language instructions is documented.
[EXTERNAL_DOWNLOADS]: The skill lists jq as a required tool in its evaluation configuration. This is a well-known, standard utility for JSON processing and is considered a safe dependency.

quality-metrics