graffiti-best-practices

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

• Third-party content exposure detected (high risk: 0.85). The required workflow explicitly fetches hosted docs/templates markdown from public URLs (e.g., https://graffiti-ui.com/base, .../templates/*) at runtime and ingests that readable text into the agent for grounding, which is outsider-authored free text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs fetching hosted Graffiti docs/templates (for example https://graffiti-ui.com/base and other https://graffiti-ui.com/* routes) at runtime to ground outputs and treat them as the required source-of-truth, so those remote URLs are fetched during runtime and directly control the agent's prompts/instructions.