mistral-common-errors

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly instructs fetching and checking third-party pages (e.g., the "curl -s https://status.mistral.ai/ | head -20" command and links to the Mistral Discord) as part of its error-handling and escalation workflow, meaning the agent would ingest external, potentially user-generated content and use it to decide retries/escalation steps.