agentos-api-knowledge

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

• Third-party content exposure detected (high risk: 0.90). The skill exposes the agent to untrusted third-party content because the Knowledge API explicitly accepts and processes uploads from URLs (references/api-patterns.md POST /knowledge/content) and the CLI/script (scripts/manage_knowledge.py) plus SKILL.md show workflows that upload, search, and use knowledge-base content which the agent reads and can act on, so arbitrary web/user-provided content could influence agent behavior.