BerriAI/litellm · error · ValueError

ANTHROPIC_API_KEY or ANTHROPIC_AUTH_TOKEN is required for Sk

Error message

ANTHROPIC_API_KEY or ANTHROPIC_AUTH_TOKEN is required for Skills API

What it means

Credential validation when building headers for the Anthropic Skills API. The transformation pulls api_key/api_base from litellm_params, resolves an auth header via AnthropicModelInfo.get_auth_header, and raises when no credential is available. The Skills API is Anthropic-direct and always requires ANTHROPIC_API_KEY or ANTHROPIC_AUTH_TOKEN.

Source

Thrown at litellm/llms/anthropic/skills/transformation.py:46

    @property
    def custom_llm_provider(self) -> LlmProviders:
        return LlmProviders.ANTHROPIC

    def validate_environment(self, headers: dict, litellm_params: GenericLiteLLMParams | None) -> dict:
        """Add Anthropic-specific headers"""
        from litellm.llms.anthropic.common_utils import AnthropicModelInfo

        # Get API key from litellm_params if available
        api_key = None
        api_base = None
        if litellm_params is not None:
            api_key = litellm_params.api_key
            api_base = litellm_params.api_base

        auth_header: Final = AnthropicModelInfo.get_auth_header(api_key, api_base)
        if auth_header is None:
            raise ValueError("ANTHROPIC_API_KEY or ANTHROPIC_AUTH_TOKEN is required for Skills API")

        headers.update(auth_header)
        headers["anthropic-version"] = "2023-06-01"

        # Add beta header for skills API
        from litellm.constants import ANTHROPIC_SKILLS_API_BETA_VERSION

        if "anthropic-beta" not in headers:
            headers["anthropic-beta"] = ANTHROPIC_SKILLS_API_BETA_VERSION
        elif isinstance(headers["anthropic-beta"], list):
            if ANTHROPIC_SKILLS_API_BETA_VERSION not in headers["anthropic-beta"]:
                headers["anthropic-beta"].append(ANTHROPIC_SKILLS_API_BETA_VERSION)
        elif isinstance(headers["anthropic-beta"], str):
            if ANTHROPIC_SKILLS_API_BETA_VERSION not in headers["anthropic-beta"]:
                headers["anthropic-beta"] = [
                    headers["anthropic-beta"],
                    ANTHROPIC_SKILLS_API_BETA_VERSION,
                ]

View on GitHub (pinned to 6c2dcb801b)

Solutions

export ANTHROPIC_API_KEY=sk-ant-... or ANTHROPIC_AUTH_TOKEN=... in the calling process.
Or set litellm_params.api_key on the skills request.
Verify with a quick env check before first use in scripts/CI.

Example fix

# before
result = skills_handler.list_skills(litellm_params=GenericLiteLLMParams())  # no key

# after
import os
os.environ["ANTHROPIC_API_KEY"] = "sk-ant-..."
result = skills_handler.list_skills(litellm_params=GenericLiteLLMParams(api_key="sk-ant-..."))

Defensive patterns

Strategy: validation

Validate before calling

import os

def skills_api_ready() -> bool:
    return bool(os.getenv("ANTHROPIC_API_KEY") or os.getenv("ANTHROPIC_AUTH_TOKEN"))

Try / catch

try:
    result = skills_endpoint_call(...)
except ValueError as e:
    if "Skills API" in str(e) and "required" in str(e):
        return http_error(503, "skills feature requires Anthropic credentials")
    raise

Prevention

Check env credentials at startup before enabling skills endpoints.
Route skills traffic through the proxy when clients lack their own keys.
Use the same secret-injection mechanism as the main Anthropic provider config.

When it happens

Trigger: Invoking a Skills API operation through the pass-through without api_key in litellm_params and without either env var set in the process. Setting only api_base (e.g. a proxy URL) without a key still fails.

Common situations: Corporate proxy setups where the key lives on the proxy but the client still sends one; fresh environments without the Anthropic secret; skills beta features exercised before credentials were configured.

Related errors

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/b6f56865d4f179bc. Report an issue: GitHub.