flowbite

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs runtime fetching of outsider-authored free text from curl -fsSL https://api.agent-ds.oday-bakkour.com/.../design.md (public web content), which can be ingested into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The SKILL.md includes a runtime curl fetch (https://api.agent-ds.oday-bakkour.com/v1/systems/flowbite/design.md) which would load a remote design.md that contains an "Agent Prompt Guide" and thus can directly control agent prompts when fetched at runtime.