speckle-errors-auth

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

• Third-party content exposure detected (high risk: 0.80). This skill's workflow (SKILL.md) explicitly instructs querying remote Speckle servers (e.g., serverInfo, project permissions, activeUser GraphQL queries and token exchange endpoints like https://app.speckle.systems or https://) and to read/interpret those responses to decide next actions (auth flows, token creation/refresh), which exposes the agent to untrusted third‑party server responses that could carry indirect prompt-injection content.