Sanitize generated pricing fields

agent_builder/skills/agent-quality-review/SKILL.md

@@ -39,6 +39,9 @@ Check these items:
   `LLMProvisioning.CALLER_PROVIDED` only when the user explicitly wants BYOK or
   caller-paid inference. In both modes, the code must use `ctx.llm` and never
   read LiteLLM/provider keys directly.
+- `Pricing` contains only `price_per_call_usd`, `caller_pays_llm`, and `notes`.
+  Reject `runtime.pricing.compute`, `runtime.pricing.total_usd`, `compute=`, or
+  `total_usd=` in `agent.py`; those are derived platform fields.
 - Any code path that builds `ChatOpenAI` checks `ctx.llm.api_key` first or
   returns a clear setup/config result before the model constructor runs.
 - Any use of DeepAgents file tools passes `backend=ctx.workspace_backend()`.