zylon-ai/private-gpt · error · ValueError

User block {i} does not end with an assistant or tool messag

Error message

User block {i} does not end with an assistant or tool message: {block[-1]}

What it means

Non-strict-mode user-block validation: a multi-message block must end with either an assistant or a tool message. It fires when the block ends with any other role — in practice a user message, meaning two consecutive user turns inside one block.

Source

Thrown at private_gpt/components/chat/processors/chat_history/memory/utils/repairs.py:149

            raise ValueError(f"User block {i} is empty.")

        # Validate the first and last messages in the block
        if block[0].role != "user":
            raise ValueError(
                f"User block {i} does not start with a user message: {block[0]}"
            )

        if len(block) == 1:
            # No more validation needed for a single user message block
            continue

        # Validate the last message in the block
        if strict and block[-1].role != "assistant":
            raise ValueError(
                f"User block {i} does not end with an assistant message: {block[-1]}"
            )
        elif not strict and block[-1].role not in ["assistant", "tool"]:
            raise ValueError(
                f"User block {i} does not end with an assistant or tool message: {block[-1]}"
            )

        # Validate the tool messages
        for j, message in enumerate(block[1:-1]):
            if message.role == "tool":
                if j == 0 or block[j].role != "assistant":
                    raise ValueError(
                        f"Tool message {j} in user block {i} does not follow an assistant message: {message}"
                    )
            elif message.role != "assistant":
                raise ValueError(
                    f"Message {j} in user block {i} is neither an assistant nor a tool: {message}"
                )


def _assert_repair_without_tools(
    chat_history: list[ChatMessage],

View on GitHub (pinned to 4a030776a3)

Solutions

  1. Merge adjacent user messages into one before repair (merge_adjacent_messages exists in the module chain)
  2. Insert a synthetic assistant acknowledgement between consecutive user turns
  3. Drop orphan user turns that never got a response if they are noise

Example fix

# before
repaired = repair_without_tools(chat_history, strict=False)

# after
from private_gpt.components.chat.processors.chat_history.memory.utils.merge import merge_adjacent_messages
repaired = repair_without_tools(merge_adjacent_messages(chat_history), strict=False)
Defensive patterns

Strategy: validation

Validate before calling

from private_gpt.components.chat.processors.chat_history.memory.utils.merge import merge_adjacent_messages
history = merge_adjacent_messages(chat_history)  # collapses consecutive user turns
blocks = get_user_blocks(history)
assert all(b[-1].role in ("assistant", "tool") for b in blocks if len(b) > 1)

Type guard

def blocks_end_valid_nonstrict(history: list[ChatMessage]) -> bool:
    return all(b[-1].role in ("assistant", "tool") for b in get_user_blocks(history) if len(b) > 1)

Try / catch

try:
    repaired = repair_without_tools(chat_history, strict=False)
except ValueError as e:
    if "does not end with an assistant or tool message" in str(e):
        repaired = repair_without_tools(merge_adjacent_messages(chat_history), strict=False)
    else:
        raise

Prevention

When it happens

Trigger: Calling repair_*(strict=False) on history where a user block's last message is a user message (user sent multiple messages without an assistant reply between them).

Common situations: Chat front-ends that let users fire messages rapidly; batch-loaded transcripts with missing assistant responses; message merging not applied before validation.

Related errors

AI-assisted analysis of zylon-ai/private-gpt@4a030776a3 (2026-08-15). Data as JSON: /api/errors/21f90740e7c39ea1. Report an issue: GitHub.