cipherstash-knowledge

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly documents runtime fetching of processed markdown for LLM consumption (e.g., https://cipherstash.com/docs/llms-full.txt), which would be retrieved at runtime and injected into the model context to control prompts, and the skill depends on that content.