trello

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

• Third-party content exposure detected (high risk: 0.75). The skill uses Trello REST API calls via curl and pipes the HTTP response through jq, so at runtime the agent ingests Trello board/list/card content (e.g., card names/descriptions/checklist items) authored by other Trello users/outsiders into the LLM context.