hono

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

• Third-party content exposure detected (high risk: 0.80). The skill's documentation and examples (e.g., references/context-and-helpers.md showing app.post('/api/items', async (c) => { const body = await c.req.json(); ... }), references/validation.md showing zValidator and c.req.valid('json'), and middleware examples reading Authorization headers) demonstrate the agent ingesting and acting on arbitrary, user-provided HTTP request content (JSON, form data, headers, query params), which is untrusted third-party input that can materially influence behavior.