feat: add max_budget_per_task configuration to control task cost (#2070)

* feat: add max_budget_per_task configuration to control task cost

* Fix test_arg_parser.py

* Use the config.max_budget_per_task as default value

* Add max_budget_per_task to core/main.py as well

* Update opendevin/controller/agent_controller.py

Co-authored-by: Boxuan Li <liboxuan@connect.hku.hk>

---------

Co-authored-by: Boxuan Li <liboxuan@connect.hku.hk>

tests/unit/test_arg_parser.py

@@ -10,8 +10,8 @@ def test_help_message(capsys):
     captured = capsys.readouterr()
     expected_help_message = """
 usage: pytest [-h] [-d DIRECTORY] [-t TASK] [-f FILE] [-c AGENT_CLS]
-              [-m MODEL_NAME] [-i MAX_ITERATIONS] [-n MAX_CHARS]
-              [--eval-output-dir EVAL_OUTPUT_DIR]
+              [-m MODEL_NAME] [-i MAX_ITERATIONS] [-b MAX_BUDGET_PER_TASK]
+              [-n MAX_CHARS] [--eval-output-dir EVAL_OUTPUT_DIR]
               [--eval-n-limit EVAL_N_LIMIT]
               [--eval-num-workers EVAL_NUM_WORKERS] [--eval-note EVAL_NOTE]
               [-l LLM_CONFIG]
@@ -31,6 +31,9 @@ options:
                         The (litellm) model name to use
   -i MAX_ITERATIONS, --max-iterations MAX_ITERATIONS
                         The maximum number of iterations to run the agent
+  -b MAX_BUDGET_PER_TASK, --max-budget-per-task MAX_BUDGET_PER_TASK
+                        The maximum budget allowed per task, beyond which the
+                        agent will stop.
   -n MAX_CHARS, --max-chars MAX_CHARS
                         The maximum number of characters to send to and
                         receive from LLM per task

tests/unit/test_config.py

@@ -285,3 +285,20 @@ def test_api_keys_repr_str():
             assert (
                 'token' not in attr_name.lower() or 'tokens' in attr_name.lower()
             ), f"Unexpected attribute '{attr_name}' contains 'token' in AppConfig"
+
+
+def test_max_iterations_and_max_budget_per_task_from_toml(temp_toml_file):
+    temp_toml = """
+[core]
+max_iterations = 100
+max_budget_per_task = 4.0
+"""
+
+    config = AppConfig()
+    with open(temp_toml_file, 'w') as f:
+        f.write(temp_toml)
+
+    load_from_toml(config, temp_toml_file)
+
+    assert config.max_iterations == 100
+    assert config.max_budget_per_task == 4.0