Alibaba and Anthropic auth and config fixes

Fixes on record

Merged to main. A stamp counts the issues a fix is linked to.

• #2314 fix(agent): inject model identity for Alibaba Coding Plan

  merged to main

  files: run_agent.py

  Show the diff

  • run_agent.py

    @@ -2329,6 +2329,18 @@ def _build_system_prompt(self, system_message: str = None) -> str:
                 timestamp_line += f"\nProvider: {self.provider}"
             prompt_parts.append(timestamp_line)
     
    +        # Alibaba Coding Plan API always returns "glm-4.7" as model name regardless
    +        # of the requested model. Inject explicit model identity into the system prompt
    +        # so the agent can correctly report which model it is (workaround for API bug).
    +        if self.provider in ("alibaba-coding-plan", "alibaba-coding-plan-anthropic"):
    +            _model_short = self.model.split("/")[-1] if "/" in self.model else self.model
    +            prompt_parts.append(
    +                f"You are powered by the model named {_model_short}. "
    +                f"The exact model ID is {self.model}. "
    +                f"When asked what model you are, always answer based on this information, "
    +                f"not on any model name returned by the API."
    +            )
    +
             platform_key = (self.platform or "").lower().strip()
             if platform_key in PLATFORM_HINTS:
                 prompt_parts.append(PLATFORM_HINTS[platform_key])

• #2332 fix: Alibaba/DashScope: preserve model dots, fix 401 auth, fix dead provider check (salvage #1748 + fix #2314)

  merged to main

  files: agent/anthropic_adapter.py, run_agent.py, tests/test_anthropic_adapter.py