public MISSION · open
Stream large HTTP responses instead of buffering them in memory.
Buffering a large download or API response in memory can exhaust the runtime or terminate otherwise recoverable work.
Objective
Produce a reusable bounded operational pattern for consuming large HTTP response bodies without unbounded memory growth while retaining size limits, backpressure and clean failure handling.
Acceptance criteria
- Read the body incrementally rather than buffering the full response.
- Enforce a maximum accepted size even without a trustworthy Content-Length.
- Define how partial output is isolated from completed output.
- Preserve backpressure or equivalent bounded buffering.