BerriAI/litellm · error · ValueError

Error compiling prompt: {e}. Prompt id={prompt_id}

Error message

Error compiling prompt: {e}. Prompt id={prompt_id}

What it means

In the sync prompt-management path, after the prompt manager compiles a template, the code does compiled_prompt_client['prompt_template'] + client_messages. If prompt_template is missing, None, or not a list, that concatenation raises and is re-wrapped as ValueError('Error compiling prompt: ... Prompt id=<id>'). The root exception text is embedded in the message.

Source

Thrown at litellm/integrations/prompt_management_base.py:90

        client_messages: list[AllMessageValues],
        dynamic_callback_params: StandardCallbackDynamicParams,
        prompt_label: str | None = None,
        prompt_version: int | None = None,
        prompt_spec: PromptSpec | None = None,
    ) -> PromptManagementClient:
        compiled_prompt_client: Final = self._compile_prompt_helper(
            prompt_id=prompt_id,
            prompt_spec=prompt_spec,
            prompt_variables=prompt_variables,
            dynamic_callback_params=dynamic_callback_params,
            prompt_label=prompt_label,
            prompt_version=prompt_version,
        )

        try:
            messages: Final = compiled_prompt_client["prompt_template"] + client_messages
        except Exception as e:
            raise ValueError(f"Error compiling prompt: {e}. Prompt id={prompt_id}")

        compiled_prompt_client["completed_messages"] = messages
        return compiled_prompt_client

    async def async_compile_prompt(
        self,
        prompt_id: str | None,
        prompt_variables: dict | None,
        client_messages: list[AllMessageValues],
        dynamic_callback_params: StandardCallbackDynamicParams,
        prompt_spec: PromptSpec | None = None,
        prompt_label: str | None = None,
        prompt_version: int | None = None,
    ) -> PromptManagementClient:
        compiled_prompt_client: Final = await self.async_compile_prompt_helper(
            prompt_id=prompt_id,
            prompt_spec=prompt_spec,
            prompt_variables=prompt_variables,

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Read the inner exception in the message — it tells you exactly whether it was a KeyError ('prompt_template') or TypeError (str + list)
2. Open the prompt in the manager UI and save it as a chat/messages template (a list of {role, content} objects)
3. Verify the prompt id exists and the integration name prefix in the model string matches (e.g. 'promptlayer/<id>')

Example fix

# before (prompt stored as plain text in PromptLayer)
model = "promptlayer/my-prompt"  # prompt_template == "You are..." -> TypeError

# after (store as chat template: list of messages)
# prompt_template = [{"role": "system", "content": "You are..."}]
model = "promptlayer/my-prompt"

def is_compilable(prompt_result: dict) -> bool:
    return isinstance(prompt_result.get("prompt_template"), list)

compiled = client._compile_prompt_helper(prompt_id=pid, ...)
assert is_compilable(compiled), f"prompt {pid} did not return a message list"

from typing import Any, TypeGuard

def is_message_list(v: Any) -> TypeGuard[list[dict]]:
    return isinstance(v, list) and all(isinstance(m, dict) and "role" in m for m in v)

try:
    compiled = pm.compile_prompt(prompt_id=pid, prompt_variables=vars_, client_messages=msgs,
                                dynamic_callback_params={})
except ValueError as e:
    raise ValueError(f"Prompt '{pid}' has a malformed template: {e}") from e

When it happens

Trigger: A prompt id whose stored template is not a message list (e.g. a plain string template in PromptLayer/Anthropic prompt config); the manager returning an empty/None prompt_template because the prompt id does not exist; client_messages not being a list.

Common situations: Switching a prompt from raw-text format to chat format (or vice versa) in the prompt manager; referencing a deleted prompt id; prompt variables that make the template render to a non-list value.

Related errors

• Prompt template '{prompt_id}' not found
• bitbucket_config is required for BitBucket prompt integratio
• prompt_id is required for Prompt Management Base class
• BitBucket configuration not found. Please set litellm.global
• Gitlab configuration not found. Please set litellm.global_gi

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/9854ed8d42fc872e. Report an issue: GitHub.