prompt-caching skills

This skill is a prompt caching system designed to reduce LLM costs. It caches prompt prefixes, full responses, and semantic similarity matches.

Source: Workspace import

Originally ingested from a local workspace copy.

Automated result: Trusted

Current public label: Trusted

The skill's documentation-focused structure means there is little executable code to analyze, and no runtime evidence was available.

Human review: none yet

The current public label is still relying on automation. A human has not weighed in yet.

Severity mix: 1 low