Compare commits

...
Author SHA1 Message Date
aakashb95 5cdd2bc14c fix(captain): mark discarded generations 2026-07-22 23:55:13 +05:30
aakashb95 8c6b70670c fix(captain): discard stale v2 responses 2026-07-22 23:21:16 +05:30
10 changed files with 207 additions and 25 deletions
@@ -7,16 +7,19 @@ class Captain::Conversation::ResponseBuilderJob < ApplicationJob
retry_on ActiveStorage::FileNotFoundError, attempts: 3, wait: 2.seconds
retry_on Faraday::BadRequestError, attempts: 3, wait: 2.seconds
def perform(conversation, assistant)
def perform(conversation, assistant, trigger_message_id = nil)
@conversation = conversation
@inbox = conversation.inbox
@assistant = assistant
@trigger_message_id = trigger_message_id if captain_v2_enabled?
return unless conversation_pending?
Current.executed_by = @assistant
if captain_v2_enabled?
return unless trigger_message_current?
generate_response_with_v2
else
generate_and_process_response
@@ -45,11 +48,18 @@ class Captain::Conversation::ResponseBuilderJob < ApplicationJob
end
def generate_response_with_v2
runner_service = Captain::Assistant::AgentRunnerService.new(assistant: @assistant, conversation: @conversation)
runner_service = Captain::Assistant::AgentRunnerService.new(
assistant: @assistant,
conversation: @conversation,
trigger_message_id: @trigger_message_id
)
message_history = Captain::Conversation::MessageHistoryBuilderService.new(conversation: @conversation).perform
@response = runner_service.generate_response(message_history: message_history)
@run_result = runner_service.last_run_result
return if runner_service.response_discarded?
return unless trigger_message_current?
process_response
end
@@ -77,16 +87,24 @@ class Captain::Conversation::ResponseBuilderJob < ApplicationJob
process_v1_handoff
elsif conversation_pending?
message = nil
ActiveRecord::Base.transaction do
message = create_messages
Rails.logger.info("[CAPTAIN][ResponseBuilderJob] Incrementing response usage for #{account.id}")
account.increment_response_usage
end
capture_assistant_session(result_message: message, credits_consumed: 1.0)
process_standard_response
end
end
def process_standard_response
message = nil
ActiveRecord::Base.transaction do
next unless trigger_message_current?
message = create_messages
Rails.logger.info("[CAPTAIN][ResponseBuilderJob] Incrementing response usage for #{account.id}")
account.increment_response_usage
end
return unless message
capture_assistant_session(result_message: message, credits_consumed: 1.0)
end
def v1_handoff_requested?
legacy_v1_handoff_token? || classifier_v1_handoff_requested?
end
@@ -117,6 +135,8 @@ class Captain::Conversation::ResponseBuilderJob < ApplicationJob
end
def process_v2_handoff
return unless trigger_message_current?
# HandoffTool already ran bot_handoff! + OOO inside the agent loop. Preserve
# waiting_since so this message doesn't clear the timestamp it left in place.
I18n.with_locale(@assistant.account.locale) do
@@ -152,7 +172,7 @@ class Captain::Conversation::ResponseBuilderJob < ApplicationJob
@response ||= {}
@response['action_source'] ||= 'error'
@response['action_reason'] ||= error_action_reason(error)
process_v1_handoff if conversation_pending?
process_v1_handoff if conversation_pending? && (!captain_v2_enabled? || trigger_message_current?)
true
end
@@ -181,4 +201,14 @@ class Captain::Conversation::ResponseBuilderJob < ApplicationJob
status = Conversation.uncached { Conversation.where(id: @conversation.id).pick(:status) }
status == 'pending' || status == Conversation.statuses[:pending]
end
def trigger_message_current?
return true if @trigger_message_id.blank?
latest_incoming_message_id = Conversation.uncached do
@conversation.messages.incoming.maximum(:id)
end
latest_incoming_message_id == @trigger_message_id
end
end
@@ -9,11 +9,12 @@ class Captain::Assistant::AgentRunnerService
attr_reader :last_run_result
def initialize(assistant:, conversation: nil, callbacks: {}, source: nil)
def initialize(assistant:, conversation: nil, callbacks: {}, source: nil, trigger_message_id: nil)
@assistant = assistant
@conversation = conversation
@callbacks = callbacks
@source = source
@trigger_message_id = trigger_message_id
@handoff_tool_called = false
end
@@ -31,6 +32,10 @@ class Captain::Assistant::AgentRunnerService
error_response(e.message)
end
def response_discarded?
@response_discarded == true
end
private
def build_context(message_history)
@@ -151,10 +156,9 @@ class Captain::Assistant::AgentRunnerService
track_handoff_usage(tool_name, handoff_tool_name, context_wrapper)
end
if ChatwootApp.otel_enabled?
runner.on_run_complete do |_agent_name, _result, context_wrapper|
write_credits_used_metadata(context_wrapper)
end
runner.on_run_complete do |_agent_name, _result, context_wrapper|
@response_discarded = trigger_message_stale?
write_run_metadata(context_wrapper) if ChatwootApp.otel_enabled?
end
runner
end
@@ -169,11 +173,22 @@ class Captain::Assistant::AgentRunnerService
@handoff_tool_called = true
end
def write_credits_used_metadata(context_wrapper)
def write_run_metadata(context_wrapper)
root_span = context_wrapper&.context&.dig(:__otel_tracing, :root_span)
return unless root_span
root_span.set_attribute(format(ATTR_LANGFUSE_METADATA, 'credit_used'), @handoff_tool_called ? 'false' : 'true')
root_span.set_attribute(format(ATTR_LANGFUSE_METADATA, 'discarded'), response_discarded?.to_s)
root_span.set_attribute(format(ATTR_LANGFUSE_METADATA, 'credit_used'), (!@handoff_tool_called && !response_discarded?).to_s)
end
def trigger_message_stale?
return false if @trigger_message_id.blank? || @conversation.blank?
latest_incoming_message_id = Conversation.uncached do
@conversation.messages.incoming.maximum(:id)
end
latest_incoming_message_id != @trigger_message_id
end
def runner
@@ -13,7 +13,8 @@ class Captain::Assistant::InstrumentationAttributeProvider
def generation_attributes(_context_wrapper, _chat, message)
{
format(ATTR_LANGFUSE_OBSERVATION_METADATA, 'generation_stage') => generation_stage(message)
format(ATTR_LANGFUSE_OBSERVATION_METADATA, 'generation_stage') => generation_stage(message),
format(ATTR_LANGFUSE_OBSERVATION_METADATA, 'discarded') => @service.send(:trigger_message_stale?).to_s
}
end
@@ -23,6 +23,7 @@ module Captain::Assistant::RunnerStateHelper
timezone: @conversation&.inbox&.timezone.presence || 'UTC'
}
state[:source] = @source if @source.present?
state[:trigger_message_id] = @trigger_message_id if @trigger_message_id.present?
build_conversation_state(state) if @conversation
state
@@ -31,6 +31,7 @@ module Enterprise::MessageTemplates::HookExecutionService
def schedule_captain_response
job_args = [conversation, conversation.inbox.captain_assistant]
job_args << message.id if conversation.account.feature_enabled?('captain_integration_v2')
if message.attachments.blank?
Captain::Conversation::ResponseBuilderJob.perform_later(*job_args)
@@ -5,6 +5,7 @@ class Captain::Tools::HandoffTool < Captain::Tools::BasePublicTool
def perform(tool_context, reason: nil)
conversation = find_conversation(tool_context.state)
return 'Conversation not found' unless conversation
return 'Handoff skipped because a newer customer message arrived' unless trigger_message_current?(tool_context.state, conversation)
# Log the handoff with reason
log_tool_usage('tool_handoff', {
@@ -23,6 +24,17 @@ class Captain::Tools::HandoffTool < Captain::Tools::BasePublicTool
private
def trigger_message_current?(state, conversation)
trigger_message_id = state&.dig(:trigger_message_id)
return true if trigger_message_id.blank?
latest_incoming_message_id = Conversation.uncached do
conversation.messages.incoming.maximum(:id)
end
latest_incoming_message_id == trigger_message_id
end
def trigger_handoff(tool_context, conversation, reason)
# post the reason as a private note
note = conversation.messages.create!(
@@ -13,6 +13,7 @@ RSpec.describe Captain::Conversation::ResponseBuilderJob, type: :job do
let(:mock_action_classifier_service) { instance_double(Captain::Llm::AssistantActionClassifierService) }
let(:mock_false_promise_service) { instance_double(Captain::Llm::AssistantFalsePromiseService) }
let(:assistant_model) { Llm::Models.default_model_for('assistant') }
let(:trigger_message) { conversation.messages.find_by!(content: 'Hello') }
before do
create(:message, conversation: conversation, content: 'Hello', message_type: :incoming)
@@ -23,6 +24,7 @@ RSpec.describe Captain::Conversation::ResponseBuilderJob, type: :job do
allow(Captain::Assistant::AgentRunnerService).to receive(:new).and_return(mock_agent_runner_service)
allow(mock_agent_runner_service).to receive(:generate_response).and_return({ 'response' => 'Hey, welcome to Captain V2' })
allow(mock_agent_runner_service).to receive(:last_run_result).and_return(nil)
allow(mock_agent_runner_service).to receive(:response_discarded?).and_return(false)
allow(Captain::Llm::AssistantActionClassifierService).to receive(:new).and_return(mock_action_classifier_service)
allow(mock_action_classifier_service).to receive(:classify).and_return({ 'action' => 'continue' })
allow(Captain::Llm::AssistantFalsePromiseService).to receive(:new).and_return(mock_false_promise_service)
@@ -358,14 +360,28 @@ RSpec.describe Captain::Conversation::ResponseBuilderJob, type: :job do
it 'uses Captain::Assistant::AgentRunnerService' do
expect(Captain::Assistant::AgentRunnerService).to receive(:new).with(
assistant: assistant,
conversation: conversation
conversation: conversation,
trigger_message_id: trigger_message.id
)
expect(Captain::Llm::AssistantChatService).not_to receive(:new)
described_class.perform_now(conversation, assistant)
described_class.perform_now(conversation, assistant, trigger_message.id)
expect(conversation.messages.last.content).to eq('Hey, welcome to Captain V2')
end
it 'discards a response when a newer message arrives during generation' do
allow(mock_agent_runner_service).to receive(:generate_response) do
create(:message, conversation: conversation, content: 'New context', message_type: :incoming)
{ 'response' => 'Stale response', 'handoff_tool_called' => false }
end
allow(mock_agent_runner_service).to receive(:response_discarded?).and_return(true)
described_class.perform_now(conversation, assistant, trigger_message.id)
expect(conversation.messages.outgoing.count).to eq(0)
expect(account.reload.usage_limits[:captain][:responses][:consumed]).to eq(0)
end
it 'passes message history with resolution markers to agent runner service' do
same_second = Time.current.change(usec: 0)
conversation.messages.find_by!(content: 'Hello').update!(created_at: same_second, updated_at: same_second)
@@ -28,6 +28,37 @@ RSpec.describe Captain::Tools::HandoffTool, type: :model do
describe '#perform' do
context 'when conversation exists' do
context 'with a trigger message id' do
let(:trigger_message) do
create(:message, conversation: conversation, account: account, inbox: inbox, message_type: :incoming, created_at: same_second)
end
let(:same_second) { Time.current.change(usec: 0) }
let(:tool_context) do
Struct.new(:state).new({ conversation: { id: conversation.id }, trigger_message_id: trigger_message.id })
end
it 'hands off when the trigger message is still the latest message' do
trigger_message
expect do
result = tool.perform(tool_context, reason: 'Customer needs specialized support')
expect(result).to include('Conversation handed off')
end.to change(Message, :count).by(1)
end
it 'skips the handoff when a newer message has arrived' do
trigger_message
conversation.update!(status: :pending)
create(:message, conversation: conversation, account: account, inbox: inbox, message_type: :incoming, created_at: same_second)
expect do
result = tool.perform(tool_context, reason: 'Customer needs specialized support')
expect(result).to eq('Handoff skipped because a newer customer message arrived')
end.not_to change(Message, :count)
expect(conversation.reload.status).to eq('pending')
end
end
context 'with reason provided' do
it 'creates a private note with reason and hands off conversation' do
reason = 'Customer needs specialized support'
@@ -52,6 +52,12 @@ RSpec.describe Captain::Assistant::AgentRunnerService do
expect(service.instance_variable_get(:@callbacks)).to eq(callbacks)
end
it 'accepts the trigger message id' do
service = described_class.new(assistant: assistant, conversation: conversation, trigger_message_id: 123)
expect(service.instance_variable_get(:@trigger_message_id)).to eq(123)
end
end
describe '#generate_response' do
@@ -99,6 +105,18 @@ RSpec.describe Captain::Assistant::AgentRunnerService do
service.generate_response(message_history: message_history)
end
it 'adds the trigger message id to the runner state' do
service = described_class.new(assistant: assistant, conversation: conversation, trigger_message_id: 123)
expect(mock_runner).to receive(:run).with(
'I need help with my account',
context: hash_including(state: hash_including(trigger_message_id: 123)),
max_turns: 10
)
service.generate_response(message_history: message_history)
end
context 'when the latest user message is multimodal' do
let(:multimodal_message_history) do
[
@@ -446,6 +464,7 @@ RSpec.describe Captain::Assistant::AgentRunnerService do
attributes = provider.generation_attributes(nil, nil, message)
expect(attributes['langfuse.observation.metadata.generation_stage']).to eq('final_response')
expect(attributes['langfuse.observation.metadata.discarded']).to eq('false')
end
it 'marks tool call generations separately from final responses' do
@@ -456,6 +475,18 @@ RSpec.describe Captain::Assistant::AgentRunnerService do
expect(attributes['langfuse.observation.metadata.generation_stage']).to eq('tool_call')
end
it 'marks a generation as discarded when a newer message has arrived' do
trigger_message = create(:message, conversation: conversation, message_type: :incoming)
runner_service = described_class.new(assistant: assistant, conversation: conversation, trigger_message_id: trigger_message.id)
attribute_provider = Captain::Assistant::InstrumentationAttributeProvider.new(runner_service)
message = instance_double(RubyLLM::Message, tool_calls: {})
create(:message, conversation: conversation, message_type: :incoming)
attributes = attribute_provider.generation_attributes(nil, nil, message)
expect(attributes['langfuse.observation.metadata.discarded']).to eq('true')
end
end
describe '#build_state' do
@@ -574,6 +605,7 @@ RSpec.describe Captain::Assistant::AgentRunnerService do
tool_complete_callback.call(Captain::Tools::HandoffTool.new(assistant).name, 'ok', context_wrapper)
expect(root_span).to receive(:set_attribute).with('langfuse.trace.metadata.discarded', 'false')
expect(root_span).to receive(:set_attribute).with('langfuse.trace.metadata.credit_used', 'false')
run_complete_callback.call('assistant', nil, context_wrapper)
end
@@ -588,6 +620,7 @@ RSpec.describe Captain::Assistant::AgentRunnerService do
tool_complete_callback = block
runner
end
allow(runner).to receive(:on_run_complete).and_return(runner)
service.send(:add_usage_metadata_callback, runner)
@@ -599,15 +632,24 @@ RSpec.describe Captain::Assistant::AgentRunnerService do
expect(context_wrapper.context[:captain_v2_handoff_tool_called]).to be true
end
it 'does not register OTEL run callback when OTEL is disabled' do
service = described_class.new(assistant: assistant, conversation: conversation)
it 'tracks discarded responses when OTEL is disabled' do
trigger_message = create(:message, conversation: conversation, message_type: :incoming)
service = described_class.new(assistant: assistant, conversation: conversation, trigger_message_id: trigger_message.id)
runner = instance_double(Agents::AgentRunner)
run_complete_callback = nil
allow(ChatwootApp).to receive(:otel_enabled?).and_return(false)
allow(runner).to receive(:on_tool_complete).and_return(runner)
expect(runner).not_to receive(:on_run_complete)
allow(runner).to receive(:on_run_complete) do |&block|
run_complete_callback = block
runner
end
service.send(:add_usage_metadata_callback, runner)
create(:message, conversation: conversation, message_type: :incoming)
run_complete_callback.call('assistant', nil, Struct.new(:context).new({}))
expect(service.response_discarded?).to be true
end
it 'sets credit_used=true when handoff tool is not used' do
@@ -629,9 +671,38 @@ RSpec.describe Captain::Assistant::AgentRunnerService do
service.send(:add_usage_metadata_callback, runner)
expect(root_span).to receive(:set_attribute).with('langfuse.trace.metadata.discarded', 'false')
expect(root_span).to receive(:set_attribute).with('langfuse.trace.metadata.credit_used', 'true')
run_complete_callback.call('assistant', nil, context_wrapper)
end
it 'marks the trace discarded and does not use credit when a newer message arrived' do
trigger_message = create(:message, conversation: conversation, message_type: :incoming)
service = described_class.new(assistant: assistant, conversation: conversation, trigger_message_id: trigger_message.id)
runner = instance_double(Agents::AgentRunner)
run_complete_callback = nil
span_class = Class.new do
def set_attribute(*); end
end
root_span = instance_double(span_class)
context_wrapper = Struct.new(:context).new({ __otel_tracing: { root_span: root_span } })
allow(ChatwootApp).to receive(:otel_enabled?).and_return(true)
allow(runner).to receive(:on_tool_complete).and_return(runner)
allow(runner).to receive(:on_run_complete) do |&block|
run_complete_callback = block
runner
end
service.send(:add_usage_metadata_callback, runner)
create(:message, conversation: conversation, message_type: :incoming)
expect(root_span).to receive(:set_attribute).with('langfuse.trace.metadata.discarded', 'true')
expect(root_span).to receive(:set_attribute).with('langfuse.trace.metadata.credit_used', 'false')
run_complete_callback.call('assistant', nil, context_wrapper)
expect(service.response_discarded?).to be true
end
end
describe 'constants' do
@@ -22,9 +22,13 @@ RSpec.describe MessageTemplates::HookExecutionService do
end
it 'schedules captain response job for incoming messages on pending conversations' do
expect(Captain::Conversation::ResponseBuilderJob).to receive(:perform_later).with(conversation, assistant)
allow(account).to receive(:feature_enabled?).and_call_original
allow(account).to receive(:feature_enabled?).with('captain_integration_v2').and_return(true)
allow(Captain::Conversation::ResponseBuilderJob).to receive(:perform_later)
create(:message, conversation: conversation, message_type: :incoming, account: account)
message = create(:message, conversation: conversation, message_type: :incoming, account: account)
expect(Captain::Conversation::ResponseBuilderJob).to have_received(:perform_later).with(conversation, assistant, message.id)
end
end