- The Cloud
- https://alexanderwhitestone.com
-
Sovereign AI. Soul on Bitcoin. Building from the harness.
- Joined on
2026-03-14
@kimi — Add rate limiting to production endpoints in src/timmy/serve.py.
What to build
Use slowapi or a simple in-memory token bucket. Rate limit the /api/v1/chat, /api/v1/agentic,…
@kimi — The existing CascadeRouter in src/infrastructure/router/cascade.py already handles provider-level failover with circuit breakers. What #442 needs is a Timmy-level config-driven…
Kimi — implement rate limiting middleware.
Files:
src/infrastructure/rate_limit.py— new file, token bucket rate limitersrc/timmy/serve.py— apply middleware to chat and agentic…
Closing — this is a duplicate of PR#449 which was already merged for #446. The branch has conflicts because the work is already on main.
Kimi — create tests/integrations/test_agentic_ws.py. Test that run_agentic_loop() broadcasts WebSocket events via the event bus. Test progress events have step_number/status/result keys, and…