666ghj/MiroFish · critical · ValueError

LLM_API_KEY 未配置

Error message

LLM_API_KEY 未配置

What it means

Raised in OasisProfileGenerator.__init__: the constructor takes an optional api_key but falls back to Config.LLM_API_KEY, and if both are empty it raises ValueError('LLM_API_KEY 未配置') before constructing the OpenAI client. This is a bootstrap-time configuration check — the LLM client for profile generation cannot be created without credentials.

Source

Thrown at backend/app/services/oasis_profile_generator.py:256

    GROUP_ENTITY_TYPES = [
        "university", "governmentagency", "organization", "ngo", 
        "mediaoutlet", "company", "institution", "group", "community"
    ]
    
    def __init__(
        self, 
        api_key: Optional[str] = None,
        base_url: Optional[str] = None,
        model_name: Optional[str] = None,
        zep_api_key: Optional[str] = None,
        graph_id: Optional[str] = None
    ):
        self.api_key = api_key or Config.LLM_API_KEY
        self.base_url = base_url or Config.LLM_BASE_URL
        self.model_name = model_name or Config.LLM_MODEL_NAME
        
        if not self.api_key:
            raise ValueError("LLM_API_KEY 未配置")
        
        self.client = OpenAI(
            api_key=self.api_key,
            base_url=self.base_url
        )
        
        # Zep客户端用于检索丰富上下文
        self.zep_api_key = zep_api_key or Config.ZEP_API_KEY
        self.zep_client = None
        self.graph_id = graph_id
        
        if self.zep_api_key:
            try:
                self.zep_client = get_zep_client(self.zep_api_key)
            except Exception as e:
                logger.warning(f"Zep客户端初始化失败: {e}")
    
    def generate_profile_from_entity(

View on GitHub (pinned to b5b53acc57)

Solutions

  1. Set LLM_API_KEY in the backend environment (.env, docker-compose environment block, or export) and restart.
  2. Or pass api_key explicitly when constructing OasisProfileGenerator.
  3. Verify the key name matches Config.LLM_API_KEY exactly and that the env file is actually loaded where the process starts.
  4. For tests, patch Config.LLM_API_KEY or inject a dummy key so construction succeeds.

Example fix

# before
gen = OasisProfileGenerator()  # ValueError: LLM_API_KEY 未配置
# after
# .env
LLM_API_KEY=sk-...
LLM_BASE_URL=https://api.example.com/v1
gen = OasisProfileGenerator()
Defensive patterns

Strategy: validation

Validate before calling

from app.config import Config
assert Config.LLM_API_KEY, 'LLM_API_KEY missing: set it in .env before starting the backend'

Try / catch

try:
    gen = OasisProfileGenerator()
except ValueError as e:
    if 'LLM_API_KEY' in str(e):
        raise SystemExit('Configure LLM_API_KEY in the environment and restart') from e
    raise

Prevention

When it happens

Trigger: Instantiating OasisProfileGenerator (directly or via the profile-generation flow) when the LLM_API_KEY environment variable is unset/empty and no api_key argument is passed. Common in fresh deployments, CI, or containers that don't load the .env file.

Common situations: Missing entry in .env / docker-compose environment; backend started without loading dotenv; typo in the variable name (LLM_APIKEY, LLM_KEY); deploying to a new environment and forgetting to port secrets; running tests that construct the service without patching Config.

Related errors

AI-assisted analysis of 666ghj/MiroFish@b5b53acc57 (2026-08-14). Data as JSON: /api/errors/06f62085dbf93002. Report an issue: GitHub.