Repository navigation
fix(OMN-11946): enforce LLM contract overlay boundaries - #839
Conversation
|
Warning Review limit reached
Your plan includes 5 reviews of capacity. Refill in 55 minutes and 6 seconds. Your organization has run out of usage credits. Purchase more in the billing tab. ⌛ How to resolve this issue?After more review capacity refills, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than trial, open-source, and free plans. In all cases, review capacity refills continuously over time. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (14)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
444937a to
d86e288
Compare
|
| Rule | Description |
|---|---|
| ARCH-TOPIC-001 | No hardcoded Kafka topic strings in handler code |
| ARCH-DI-001 | No event_bus=None bypass |
| ARCH-DI-002 | No direct Handler instantiation outside workflow_runner/adapters |
| ARCH-DI-003 | No reinvented DI containers |
| ARCH-TOPIC-002 | contract.yaml topics follow `onex.{cmd |
ARCH-DI-002: Direct Handler instantiation (warning)\n\n\nsrc/omnimarket/nodes/node_generation_consumer/handlers/handler_generation_consumer.py:272: self._effect = HandlerLlmOpenaiCompatible(transport=_Transport())\n\n\nHandler classes should be wired only in workflow_runner.py or adapter modules. Direct instantiation elsewhere bypasses the contract system.\n\n
Powered by node_pr_review_bot
✅ PR Review Bot — CLEANFindings: 0
Verdict semantics
Powered by node_pr_review_bot — Qwen3-Coder-30B reviewer + DeepSeek-R1 judge (OMN-7963) |
|
| Verdict | Meaning | Blocks merge? |
|---|---|---|
passed |
No critical/error findings | No |
blocked |
CRITICAL or ERROR findings found | Yes |
degraded |
All models unavailable (infra) | No (pilot) |
Powered by omniintelligence.review_pairing.cli_review — node-based adversarial review via HandlerLlmCliSubprocess (OMN-8524)
Summary
Evidence
uv run python scripts/validation/validate_contract_overlay_boundary.py --root .uv run ruff check src/omnimarket/nodes/node_adr_segmentation_llm_effect/handlers/handler_segmentation.py src/omnimarket/nodes/node_ci_fix_effect/handlers/handler_ci_fix.py src/omnimarket/nodes/node_generation_consumer/handlers/handler_generation_consumer.py tests/nodes/node_ci_fix_effect/test_handler_ci_fix.py tests/unit/nodes/node_generation_consumer/test_handler_generation_consumer.pyuv run pytest tests/validation/test_contract_overlay_boundary.py tests/nodes/node_thread_reply_effect/test_handler_thread_reply.py src/omnimarket/nodes/node_adr_segmentation_llm_effect/tests/test_handler_segmentation.py tests/test_golden_chain_golden_chain_sweep.py tests/test_golden_chain_build_loop_events.py tests/test_node_model_router_contract_boundary.py tests/test_model_router_routes_to_local_first.py tests/test_model_router_health_check_degrades.py tests/test_model_router_publishes_degradation_event.py tests/test_model_router_authorization_and_ttl.py tests/nodes/node_ci_fix_effect/test_handler_ci_fix.py tests/unit/nodes/node_generation_consumer/test_handler_generation_consumer.py -quv run python -m omnimarket.nodes.node_golden_chain_sweep --chains routing,delegation --projected-rows ...Ticket
OMN-11946
Evidence-Source: b63923360b58fe7537c1ccafe3e2a6b064423935
Evidence-Ticket: OMN-11946