sjtu_text

Instructions

Implementation Reference

• src/sjtu_mcp/server.py:68-83 (handler)

  The main handler for the 'sjtu_text' MCP tool. It accepts a prompt, optional model/system_prompt/temperature/max_tokens, builds text messages via _build_text_messages, calls the SJTU chat API via client.chat, and extracts the response text via _extract_text.

  @mcp.tool()
  async def sjtu_text(
      prompt: str,
      model: str | None = None,
      system_prompt: str | None = None,
      temperature: float = 0.2,
      max_tokens: int | None = None,
  ) -> str:
      """Run a plain text task against the SJTU OpenAI-compatible API."""
      response = await client.chat(
          model=model or settings.default_text_model,
          messages=_build_text_messages(prompt, system_prompt),
          temperature=temperature,
          max_tokens=max_tokens,
      )
      return _extract_text(response)

• src/sjtu_mcp/server.py:68-68 (registration)

  The tool is registered with the MCP framework via the @mcp.tool() decorator on line 62 (for sjtu_models) and line 68 (for sjtu_text). The 'FastMCP' instance named 'mcp' is created on line 12 and the decorator registers the function as an MCP tool.

  @mcp.tool()

• src/sjtu_mcp/server.py:15-29 (helper)

  Helper function _extract_text parses the API response dict to extract the text content from the first choice's message. Handles both plain string and list-of-content-blocks formats.

  def _extract_text(response: dict[str, Any]) -> str:
      choices = response.get("choices", [])
      if not choices:
          return "No choices returned."
      message = choices[0].get("message", {})
      content = message.get("content", "")
      if isinstance(content, str):
          return content
      if isinstance(content, list):
          chunks: list[str] = []
          for item in content:
              if isinstance(item, dict) and item.get("type") == "text":
                  chunks.append(str(item.get("text", "")))
          return "\n".join(chunk for chunk in chunks if chunk)
      return str(content)

• src/sjtu_mcp/server.py:32-37 (helper)

  Helper function _build_text_messages constructs the messages list for a plain text chat completion request, optionally including a system prompt.

  def _build_text_messages(prompt: str, system_prompt: str | None) -> list[dict[str, Any]]:
      messages: list[dict[str, Any]] = []
      if system_prompt:
          messages.append({"role": "system", "content": system_prompt})
      messages.append({"role": "user", "content": prompt})
      return messages

• src/sjtu_mcp/server.py:70-77 (schema)

  The schema for sjtu_text is defined implicitly through the function signature parameters: prompt (str, required), model (str | None), system_prompt (str | None), temperature (float, default 0.2), max_tokens (int | None). The return type is str.

      prompt: str,
      model: str | None = None,
      system_prompt: str | None = None,
      temperature: float = 0.2,
      max_tokens: int | None = None,
  ) -> str:
      """Run a plain text task against the SJTU OpenAI-compatible API."""
      response = await client.chat(