From 3debef578143251618625c45ed14b1530373b5e6 Mon Sep 17 00:00:00 2001 From: Laine Date: Wed, 19 Mar 2025 12:48:49 -0400 Subject: [PATCH 1/2] Add clarification re: format of start and end dates, and (probably?) fix non-determinism error caused by SHOW_CONFIRM --- tools/tool_registry.py | 9 +++++---- workflows/agent_goal_workflow.py | 8 ++++---- 2 files changed, 9 insertions(+), 8 deletions(-) diff --git a/tools/tool_registry.py b/tools/tool_registry.py index 79a4225..9886435 100644 --- a/tools/tool_registry.py +++ b/tools/tool_registry.py @@ -50,6 +50,7 @@ guess_location_tool = ToolDefinition( ), ], ) + # ----- Travel use cases tools ----- search_flights_tool = ToolDefinition( name="SearchFlights", @@ -198,12 +199,12 @@ future_pto_calc_tool = ToolDefinition( ToolArgument( name="start_date", type="string", - description="Start date of proposed PTO", + description="Start date of proposed PTO, sent in the form yyyy-mm-dd", ), ToolArgument( name="end_date", type="string", - description="End date of proposed PTO", + description="End date of proposed PTO, sent in the form yyyy-mm-dd", ), ToolArgument( name="email", @@ -221,12 +222,12 @@ book_pto_tool = ToolDefinition( ToolArgument( name="start_date", type="string", - description="Start date of proposed PTO", + description="Start date of proposed PTO, sent in the form yyyy-mm-dd", ), ToolArgument( name="end_date", type="string", - description="End date of proposed PTO", + description="End date of proposed PTO, sent in the form yyyy-mm-dd", ), ToolArgument( name="email", diff --git a/workflows/agent_goal_workflow.py b/workflows/agent_goal_workflow.py index 4699e86..86324ff 100644 --- a/workflows/agent_goal_workflow.py +++ b/workflows/agent_goal_workflow.py @@ -26,11 +26,11 @@ with workflow.unsafe.imports_passed_through(): # Constants MAX_TURNS_BEFORE_CONTINUE = 250 -SHOW_CONFIRM = True show_confirm_env = os.getenv("SHOW_CONFIRM") -if show_confirm_env is not None: - if show_confirm_env == "False": - SHOW_CONFIRM = False +if show_confirm_env is not None and show_confirm_env.lower() == "false": + SHOW_CONFIRM = False +else: + SHOW_CONFIRM = True #ToolData as part of the workflow is what's accessible to the UI - see LLMResponse.jsx for example class ToolData(TypedDict, total=False): From d8a8fe44f9f28e9cc2ac4a7a4c28c033d80f62e0 Mon Sep 17 00:00:00 2001 From: Laine Date: Wed, 19 Mar 2025 15:28:20 -0400 Subject: [PATCH 2/2] Add additional hints --- tools/data/employee_pto_data.json | 2 +- tools/give_hint.py | 13 ++++++------- 2 files changed, 7 insertions(+), 8 deletions(-) diff --git a/tools/data/employee_pto_data.json b/tools/data/employee_pto_data.json index 38a6329..c5c38d6 100644 --- a/tools/data/employee_pto_data.json +++ b/tools/data/employee_pto_data.json @@ -8,7 +8,7 @@ "hrsAddedPerMonth": 8 }, { - "email": "lainecaseysmith@gmail.com", + "email": "laine@awesome.com", "currentPTOHrs": 40, "hrsAddedPerMonth": 12 } diff --git a/tools/give_hint.py b/tools/give_hint.py index a8072ad..a2fad2f 100644 --- a/tools/give_hint.py +++ b/tools/give_hint.py @@ -8,17 +8,17 @@ TREASURE_LOCATION = { } HINTS = [ - "state of Washington", - "city of Seattle", + "country of " + TREASURE_LOCATION["country"], + "state of " + TREASURE_LOCATION["state_full"], + "city of " + TREASURE_LOCATION["city"], "at a company HQ", + "The company's tech traces its roots to a project called Cadence", #thanks, Grok + "The company offers a tool that lets developers write code as if it's running forever, no matter what crashes", #thanks, Grok ] -''' Grok provided hints: -Here are additional company-specific clues about Temporal that could help players in your game guess the address (300 Lenora St, Seattle, WA) by focusing on the company itself. These are designed to be intriguing and game-friendly: - +''' Additional Grok provided hints about Temporal: "This company was founded by two engineers who previously worked on a system named after a South American river at Uber." "Their platform is all about orchestrating workflows that can survive failures—like a conductor keeping the music going." "They offer a tool that lets developers write code as if it’s running forever, no matter what crashes." -"The company’s tech traces its roots to a project called Cadence, which they took to the next level." "Their mission is tied to making distributed systems feel as simple as writing a single app." "They’ve got a knack for ‘durability’—both in their software and their growing reputation." "This outfit spun out of experiences at AWS and Uber, blending cloud and ride-sharing know-how." @@ -34,7 +34,6 @@ def give_hint(args: dict) -> dict: index = hint_total % len(HINTS) hint_text = HINTS[index] - print(f"hint_total: {hint_total}, length: {len(HINTS)}, index: {index}") hint_total = hint_total + 1 return { "hint_number": hint_total,