PromptSmith
TARGET: OpenAI o1, o3-mini & ChatGPT-4o
OpenAI o1 & o3-Mini Prompt Structurer (Anti-Loop)
Conversational filler tokens like "could you please" burn 35% of self-attention compute. PromptSmith decomposes draft instructions into high-density structured contracts in zero milliseconds.
Raw Input Draft
Natural English
High-Density Prompt Output
-38% Tokens Saved
### Problem Specification
Implement a distributed Token Bucket rate limiter middleware for FastAPI backed by Redis.
### Invariant & Boundary Criteria
- Atomic token consumption via single-round Redis Lua script (prevent race conditions).
- Strict fallback to local in-memory token bucket if Redis connection drops (fail-open).
- Response headers: X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset.
### Acceptance Verification Assertions
- Asynchronous pytest assertions validating concurrency under 100 simultaneous requests.