subagent-driven-development

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

• Third-party content exposure detected (high risk: 1.00). The skill's required workflow explicitly tells sub-agents to self-read user-generated issue descriptions and epic comments via bd show and bd comments (see context-loading.md, implementer-prompt.md, and spec/code-reviewer prompts), so arbitrary/untrusted third-party content stored in beads is ingested and can directly influence tool use and agent actions.