atlassian

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

• Third-party content exposure detected (high risk: 0.90). The skill's confluence wrapper (scripts/confluence) and SKILL.md / references (e.g., references/confluence-commands.md) explicitly fetch Confluence pages, search results, and footer comments via the Confluence REST APIs (e.g., /wiki/rest/api/search and /wiki/api/v2/pages/... used by confluence search/get/comments), which exposes user-generated, untrusted third-party content that the agent is expected to read and that can materially influence subsequent actions (pipelines show page bodies/comments being read and used to drive updates or other commands).