fusion-issue-authoring

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly fetches and interprets untrusted, user-generated content from target GitHub repositories (e.g., repository templates in .github/ISSUE_TEMPLATE/ and issue data via mcp_github::search_issues, mcp_github::list_issue_types and the provided GraphQL queries), as described in SKILL.md and references/mcp-server.md, and uses that content to decide templates, labels, duplicate detection, issue types and subsequent MCP mutation actions.