harper-best-practices

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

• Third-party content exposure detected (high risk: 0.70). The skill's required workflows instruct Resources to fetch and ingest external HTTP APIs (e.g., rules/caching.md shows a Resource fetching https://api.example.com/items/{id}) and to call external embedding services (rules/vector-indexing.md uses OpenAI/Ollama), meaning the agent will read and act on untrusted third‑party content that can influence searches, caching, and subsequent actions.