linear

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime path is calling Linear’s GraphQL API via curl and ingesting the JSON response (including outsider-authored issue/comment body text) into the agent context; e.g., Get a single issue fetches comments { nodes { body user { name } ... } } and Get a single issue also returns description.