wasp-plugin-init

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to "find and fetch" the Wasp documentation map from the public URL https://wasp.sh/llms.txt and then fetch guides from raw.githubusercontent.com (see general-wasp-knowledge.md), meaning it fetches and ingests untrusted, public third-party markdown which the agent is told to use to inform its actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent at runtime to fetch the LLM index from https://wasp.sh/llms.txt and then retrieve raw documentation markdown from raw.githubusercontent.com URLs to inject into the agent's context/knowledge, meaning external content would directly control prompts/instructions.