From 0af9900be2f0bfb63bae227f7f7c734d4769e92f Mon Sep 17 00:00:00 2001 From: Aakash Bakhle <48802744+aakashb95@users.noreply.github.com> Date: Tue, 23 Jun 2026 18:42:34 +0530 Subject: [PATCH] fix: model for false promise harness (#14824) fixes model for false promise harness --- .../captain/llm/assistant_false_promise_service.rb | 6 ++++++ .../conversation/response_builder_job_spec.rb | 14 +++++++------- 2 files changed, 13 insertions(+), 7 deletions(-) diff --git a/enterprise/app/services/captain/llm/assistant_false_promise_service.rb b/enterprise/app/services/captain/llm/assistant_false_promise_service.rb index 56d703e77..b19f3fd11 100644 --- a/enterprise/app/services/captain/llm/assistant_false_promise_service.rb +++ b/enterprise/app/services/captain/llm/assistant_false_promise_service.rb @@ -1,4 +1,6 @@ class Captain::Llm::AssistantFalsePromiseService < Llm::BaseAiService + DETECTOR_MODEL = 'gpt-5.2'.freeze + include Integrations::LlmInstrumentation include Captain::Llm::AssistantResponseInspectionHelpers @@ -35,6 +37,10 @@ class Captain::Llm::AssistantFalsePromiseService < Llm::BaseAiService private + def setup_model + @model = DETECTOR_MODEL + end + def normalize_response(parsed, raw_content) decision = parsed['decision'].to_s reason = parsed['reason'].to_s diff --git a/spec/enterprise/jobs/captain/conversation/response_builder_job_spec.rb b/spec/enterprise/jobs/captain/conversation/response_builder_job_spec.rb index 3cb6958f5..4ce37523c 100644 --- a/spec/enterprise/jobs/captain/conversation/response_builder_job_spec.rb +++ b/spec/enterprise/jobs/captain/conversation/response_builder_job_spec.rb @@ -82,7 +82,7 @@ RSpec.describe Captain::Conversation::ResponseBuilderJob, type: :job do ).and_return({ 'decision' => 'safe', 'reason' => 'safe_response', - 'model' => 'gpt-4.1' + 'model' => Captain::Llm::AssistantFalsePromiseService::DETECTOR_MODEL }) described_class.perform_now(conversation, assistant) @@ -103,12 +103,12 @@ RSpec.describe Captain::Conversation::ResponseBuilderJob, type: :job do { 'decision' => 'future_work_promise', 'reason' => 'future_check_or_investigation', - 'model' => 'gpt-4.1' + 'model' => Captain::Llm::AssistantFalsePromiseService::DETECTOR_MODEL }, { 'decision' => 'safe', 'reason' => 'asks_user_to_check_or_provide_info', - 'model' => 'gpt-4.1' + 'model' => Captain::Llm::AssistantFalsePromiseService::DETECTOR_MODEL } ) @@ -143,7 +143,7 @@ RSpec.describe Captain::Conversation::ResponseBuilderJob, type: :job do allow(mock_false_promise_service).to receive(:detect).and_return({ 'decision' => 'future_work_promise', 'reason' => 'future_check_or_investigation', - 'model' => 'gpt-4.1' + 'model' => Captain::Llm::AssistantFalsePromiseService::DETECTOR_MODEL }) described_class.perform_now(conversation, assistant) @@ -165,13 +165,13 @@ RSpec.describe Captain::Conversation::ResponseBuilderJob, type: :job do { 'decision' => 'future_work_promise', 'reason' => 'future_check_or_investigation', - 'model' => 'gpt-4.1' + 'model' => Captain::Llm::AssistantFalsePromiseService::DETECTOR_MODEL }, { 'decision' => nil, 'reason' => nil, 'error' => 'verification timeout', - 'model' => 'gpt-4.1' + 'model' => Captain::Llm::AssistantFalsePromiseService::DETECTOR_MODEL } ) @@ -192,7 +192,7 @@ RSpec.describe Captain::Conversation::ResponseBuilderJob, type: :job do allow(mock_false_promise_service).to receive(:detect).and_return({ 'decision' => 'future_work_promise', 'reason' => 'future_check_or_investigation', - 'model' => 'gpt-4.1' + 'model' => Captain::Llm::AssistantFalsePromiseService::DETECTOR_MODEL }) described_class.perform_now(conversation, assistant)