0a8a8fda54
* feat: add credit verification before agent message endpoints
Add credit verification checks to message endpoints to prevent
execution when organizations have insufficient credits.
- Add InsufficientCreditsError exception type
- Add CreditVerificationService that calls step-orchestrator API
- Add credit checks to /agents/{id}/messages endpoints
- Add credit checks to /conversations/{id}/messages endpoint
🐾 Generated with [Letta Code](https://letta.com)
Co-Authored-By: Letta <noreply@letta.com>
* surface error in ade
* do per step instead
* parallel check
* parallel to step
* small fixes
* stage publish api
* fixes
* revert unnecessary frontend changes
* insufficient credits stop reason
---------
Co-authored-by: Letta <noreply@letta.com>
59 lines
2.0 KiB
Python
59 lines
2.0 KiB
Python
from enum import Enum
|
|
from typing import Literal
|
|
|
|
from pydantic import BaseModel, Field
|
|
|
|
from letta.schemas.enums import RunStatus
|
|
|
|
|
|
class StopReasonType(str, Enum):
|
|
end_turn = "end_turn"
|
|
error = "error"
|
|
llm_api_error = "llm_api_error"
|
|
invalid_llm_response = "invalid_llm_response"
|
|
invalid_tool_call = "invalid_tool_call"
|
|
max_steps = "max_steps"
|
|
max_tokens_exceeded = "max_tokens_exceeded"
|
|
no_tool_call = "no_tool_call"
|
|
tool_rule = "tool_rule"
|
|
cancelled = "cancelled"
|
|
insufficient_credits = "insufficient_credits"
|
|
requires_approval = "requires_approval"
|
|
context_window_overflow_in_system_prompt = "context_window_overflow_in_system_prompt"
|
|
|
|
@property
|
|
def run_status(self) -> RunStatus:
|
|
if self in (
|
|
StopReasonType.end_turn,
|
|
StopReasonType.max_steps,
|
|
StopReasonType.tool_rule,
|
|
StopReasonType.requires_approval,
|
|
):
|
|
return RunStatus.completed
|
|
elif self in (
|
|
StopReasonType.error,
|
|
StopReasonType.invalid_tool_call,
|
|
StopReasonType.no_tool_call,
|
|
StopReasonType.invalid_llm_response,
|
|
StopReasonType.llm_api_error,
|
|
# Treat context/token limit exhaustion as an error state (same as llm_api_error)
|
|
StopReasonType.max_tokens_exceeded,
|
|
StopReasonType.context_window_overflow_in_system_prompt,
|
|
):
|
|
return RunStatus.failed
|
|
elif self == StopReasonType.cancelled:
|
|
return RunStatus.cancelled
|
|
elif self == StopReasonType.insufficient_credits:
|
|
return RunStatus.failed
|
|
else:
|
|
raise ValueError("Unknown StopReasonType")
|
|
|
|
|
|
class LettaStopReason(BaseModel):
|
|
"""
|
|
The stop reason from Letta indicating why agent loop stopped execution.
|
|
"""
|
|
|
|
message_type: Literal["stop_reason"] = Field("stop_reason", description="The type of the message.")
|
|
stop_reason: StopReasonType = Field(..., description="The reason why execution stopped.")
|