worma-guidelines

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs the agent to "fetch the detailed information from the corresponding docs link before answering" (including the index https://worma.js.org/llms.txt and many https://worma.js.org/... docs links), so external documentation would be fetched at runtime and could directly control the agent's prompts/behavior.