tapd-bug-analyzer

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly calls the TAPD MCP tool (mcp__mcp-server-tapd__get_bug) to fetch the TAPD Bug "description" field (user-generated/untrusted content) and then parses that content for traceIds/URLs which directly drive downstream tool calls (e.g., sls-trace-analyzer, codebase-retrieval), so third-party text can materially change agent actions.