short-alpha-general

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md runtime discipline and bootstrap/readme explicitly require using external search layers (see "Search Runtime Contract" / workspace/SEARCH_RUNTIME.md and the listed sequence "finance-mcp -> open-websearch -> zhipu -> metaso -> tavily -> brave" and "first executable search action must be open-websearch"), so the agent is expected to fetch and ingest untrusted public web content that can materially influence pipeline decisions.