planning-with-files

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (SKILL.md's "2-Action Rule" and "Read vs Write Decision Matrix") and accompanying templates/examples (examples.md and templates/findings.md) explicitly instruct the agent to perform WebSearch/view browser results and record external URLs/content, meaning the agent will ingest untrusted public web content that can influence its decisions and tool use.