hud-python 0.4.1__py3-none-any.whl → 0.4.3__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.


This version of hud-python might be problematic. Click here for more details.

Files changed (130) hide show
  1. hud/__init__.py +22 -22
  2. hud/agents/__init__.py +13 -15
  3. hud/agents/base.py +599 -599
  4. hud/agents/claude.py +373 -373
  5. hud/agents/langchain.py +261 -250
  6. hud/agents/misc/__init__.py +7 -7
  7. hud/agents/misc/response_agent.py +82 -80
  8. hud/agents/openai.py +352 -352
  9. hud/agents/openai_chat_generic.py +154 -154
  10. hud/agents/tests/__init__.py +1 -1
  11. hud/agents/tests/test_base.py +742 -742
  12. hud/agents/tests/test_claude.py +324 -324
  13. hud/agents/tests/test_client.py +363 -363
  14. hud/agents/tests/test_openai.py +237 -237
  15. hud/cli/__init__.py +617 -617
  16. hud/cli/__main__.py +8 -8
  17. hud/cli/analyze.py +371 -371
  18. hud/cli/analyze_metadata.py +230 -230
  19. hud/cli/build.py +498 -427
  20. hud/cli/clone.py +185 -185
  21. hud/cli/cursor.py +92 -92
  22. hud/cli/debug.py +392 -392
  23. hud/cli/docker_utils.py +83 -83
  24. hud/cli/init.py +280 -281
  25. hud/cli/interactive.py +353 -353
  26. hud/cli/mcp_server.py +764 -756
  27. hud/cli/pull.py +330 -336
  28. hud/cli/push.py +404 -370
  29. hud/cli/remote_runner.py +311 -311
  30. hud/cli/runner.py +160 -160
  31. hud/cli/tests/__init__.py +3 -3
  32. hud/cli/tests/test_analyze.py +284 -284
  33. hud/cli/tests/test_cli_init.py +265 -265
  34. hud/cli/tests/test_cli_main.py +27 -27
  35. hud/cli/tests/test_clone.py +142 -142
  36. hud/cli/tests/test_cursor.py +253 -253
  37. hud/cli/tests/test_debug.py +453 -453
  38. hud/cli/tests/test_mcp_server.py +139 -139
  39. hud/cli/tests/test_utils.py +388 -388
  40. hud/cli/utils.py +263 -263
  41. hud/clients/README.md +143 -143
  42. hud/clients/__init__.py +16 -16
  43. hud/clients/base.py +378 -379
  44. hud/clients/fastmcp.py +222 -222
  45. hud/clients/mcp_use.py +298 -278
  46. hud/clients/tests/__init__.py +1 -1
  47. hud/clients/tests/test_client_integration.py +111 -111
  48. hud/clients/tests/test_fastmcp.py +342 -342
  49. hud/clients/tests/test_protocol.py +188 -188
  50. hud/clients/utils/__init__.py +1 -1
  51. hud/clients/utils/retry_transport.py +160 -160
  52. hud/datasets.py +327 -322
  53. hud/misc/__init__.py +1 -1
  54. hud/misc/claude_plays_pokemon.py +292 -292
  55. hud/otel/__init__.py +35 -35
  56. hud/otel/collector.py +142 -142
  57. hud/otel/config.py +164 -164
  58. hud/otel/context.py +536 -536
  59. hud/otel/exporters.py +366 -366
  60. hud/otel/instrumentation.py +97 -97
  61. hud/otel/processors.py +118 -118
  62. hud/otel/tests/__init__.py +1 -1
  63. hud/otel/tests/test_processors.py +197 -197
  64. hud/server/__init__.py +5 -5
  65. hud/server/context.py +114 -114
  66. hud/server/helper/__init__.py +5 -5
  67. hud/server/low_level.py +132 -132
  68. hud/server/server.py +170 -166
  69. hud/server/tests/__init__.py +3 -3
  70. hud/settings.py +73 -73
  71. hud/shared/__init__.py +5 -5
  72. hud/shared/exceptions.py +180 -180
  73. hud/shared/requests.py +264 -264
  74. hud/shared/tests/test_exceptions.py +157 -157
  75. hud/shared/tests/test_requests.py +275 -275
  76. hud/telemetry/__init__.py +25 -25
  77. hud/telemetry/instrument.py +379 -379
  78. hud/telemetry/job.py +309 -309
  79. hud/telemetry/replay.py +74 -74
  80. hud/telemetry/trace.py +83 -83
  81. hud/tools/__init__.py +33 -33
  82. hud/tools/base.py +365 -365
  83. hud/tools/bash.py +161 -161
  84. hud/tools/computer/__init__.py +15 -15
  85. hud/tools/computer/anthropic.py +437 -437
  86. hud/tools/computer/hud.py +376 -376
  87. hud/tools/computer/openai.py +295 -295
  88. hud/tools/computer/settings.py +82 -82
  89. hud/tools/edit.py +314 -314
  90. hud/tools/executors/__init__.py +30 -30
  91. hud/tools/executors/base.py +539 -539
  92. hud/tools/executors/pyautogui.py +621 -621
  93. hud/tools/executors/tests/__init__.py +1 -1
  94. hud/tools/executors/tests/test_base_executor.py +338 -338
  95. hud/tools/executors/tests/test_pyautogui_executor.py +165 -165
  96. hud/tools/executors/xdo.py +511 -511
  97. hud/tools/playwright.py +412 -412
  98. hud/tools/tests/__init__.py +3 -3
  99. hud/tools/tests/test_base.py +282 -282
  100. hud/tools/tests/test_bash.py +158 -158
  101. hud/tools/tests/test_bash_extended.py +197 -197
  102. hud/tools/tests/test_computer.py +425 -425
  103. hud/tools/tests/test_computer_actions.py +34 -34
  104. hud/tools/tests/test_edit.py +259 -259
  105. hud/tools/tests/test_init.py +27 -27
  106. hud/tools/tests/test_playwright_tool.py +183 -183
  107. hud/tools/tests/test_tools.py +145 -145
  108. hud/tools/tests/test_utils.py +156 -156
  109. hud/tools/types.py +72 -72
  110. hud/tools/utils.py +50 -50
  111. hud/types.py +136 -136
  112. hud/utils/__init__.py +10 -10
  113. hud/utils/async_utils.py +65 -65
  114. hud/utils/design.py +236 -168
  115. hud/utils/mcp.py +55 -55
  116. hud/utils/progress.py +149 -149
  117. hud/utils/telemetry.py +66 -66
  118. hud/utils/tests/test_async_utils.py +173 -173
  119. hud/utils/tests/test_init.py +17 -17
  120. hud/utils/tests/test_progress.py +261 -261
  121. hud/utils/tests/test_telemetry.py +82 -82
  122. hud/utils/tests/test_version.py +8 -8
  123. hud/version.py +7 -7
  124. {hud_python-0.4.1.dist-info → hud_python-0.4.3.dist-info}/METADATA +10 -8
  125. hud_python-0.4.3.dist-info/RECORD +131 -0
  126. {hud_python-0.4.1.dist-info → hud_python-0.4.3.dist-info}/licenses/LICENSE +21 -21
  127. hud/agents/art.py +0 -101
  128. hud_python-0.4.1.dist-info/RECORD +0 -132
  129. {hud_python-0.4.1.dist-info → hud_python-0.4.3.dist-info}/WHEEL +0 -0
  130. {hud_python-0.4.1.dist-info → hud_python-0.4.3.dist-info}/entry_points.txt +0 -0
@@ -1,80 +1,82 @@
1
- from __future__ import annotations
2
-
3
- import os
4
- from typing import Literal
5
-
6
- from openai import AsyncOpenAI
7
-
8
- ResponseType = Literal["STOP", "CONTINUE"]
9
-
10
-
11
- class ResponseAgent:
12
- """
13
- An assistant that helps determine whether an agent should stop or continue
14
- based on the agent's final response message.
15
- """
16
-
17
- def __init__(self, api_key: str | None = None) -> None:
18
- self.api_key = api_key or os.environ.get("OPENAI_API_KEY")
19
- if not self.api_key:
20
- raise ValueError(
21
- "OpenAI API key must be provided or set as OPENAI_API_KEY environment variable"
22
- )
23
-
24
- self.client = AsyncOpenAI(api_key=self.api_key)
25
-
26
- self.system_prompt = """
27
- You are an assistant that helps determine the appropriate response to an agent's message.
28
-
29
- You will receive messages from an agent that is performing tasks for a user.
30
- Your job is to analyze these messages and respond with one of the following:
31
-
32
- - STOP: If the agent indicates it has successfully completed a task, even if phrased as a question
33
- like "I have entered the right values into this form. Would you like me to do anything else?"
34
- or "Here is the website. Is there any other information you need?"
35
-
36
- - CONTINUE: If the agent is asking for clarification before proceeding with a task
37
- like "I'm about to clear cookies from this website. Would you like me to proceed?"
38
- or "I've entered the right values into this form. Would you like me to continue with the rest of the task?"
39
-
40
- Respond ONLY with one of these two options.
41
- """ # noqa: E501
42
-
43
- async def determine_response(self, agent_message: str) -> ResponseType:
44
- """
45
- Determine whether the agent should stop or continue based on its message.
46
-
47
- Args:
48
- agent_message: The message from the agent
49
-
50
- Returns:
51
- ResponseType: Either "STOP" or "CONTINUE"
52
- """
53
- try:
54
- response = await self.client.chat.completions.create(
55
- model="gpt-4o",
56
- messages=[
57
- {"role": "system", "content": self.system_prompt},
58
- {
59
- "role": "user",
60
- "content": f"Agent message: {agent_message}\n\nWhat is the appropriate response?", # noqa: E501
61
- },
62
- ],
63
- temperature=0.1, # Low temperature for more deterministic responses
64
- max_tokens=5, # We only need a short response
65
- )
66
-
67
- response_text = response.choices[0].message.content
68
- if not response_text:
69
- return "CONTINUE"
70
-
71
- response_text = response_text.strip().upper()
72
-
73
- # Validate the response
74
- if "STOP" in response_text:
75
- return "STOP"
76
- else:
77
- return "CONTINUE"
78
-
79
- except Exception:
80
- return "CONTINUE" # Default to continue on error
1
+ from __future__ import annotations
2
+
3
+ import os
4
+ from typing import Literal
5
+
6
+ from openai import AsyncOpenAI
7
+
8
+ from hud.settings import settings
9
+
10
+ ResponseType = Literal["STOP", "CONTINUE"]
11
+
12
+
13
+ class ResponseAgent:
14
+ """
15
+ An assistant that helps determine whether an agent should stop or continue
16
+ based on the agent's final response message.
17
+ """
18
+
19
+ def __init__(self, api_key: str | None = None) -> None:
20
+ self.api_key = api_key or settings.openai_api_key or os.environ.get("OPENAI_API_KEY")
21
+ if not self.api_key:
22
+ raise ValueError(
23
+ "OpenAI API key must be provided or set as OPENAI_API_KEY environment variable"
24
+ )
25
+
26
+ self.client = AsyncOpenAI(api_key=self.api_key)
27
+
28
+ self.system_prompt = """
29
+ You are an assistant that helps determine the appropriate response to an agent's message.
30
+
31
+ You will receive messages from an agent that is performing tasks for a user.
32
+ Your job is to analyze these messages and respond with one of the following:
33
+
34
+ - STOP: If the agent indicates it has successfully completed a task, even if phrased as a question
35
+ like "I have entered the right values into this form. Would you like me to do anything else?"
36
+ or "Here is the website. Is there any other information you need?"
37
+
38
+ - CONTINUE: If the agent is asking for clarification before proceeding with a task
39
+ like "I'm about to clear cookies from this website. Would you like me to proceed?"
40
+ or "I've entered the right values into this form. Would you like me to continue with the rest of the task?"
41
+
42
+ Respond ONLY with one of these two options.
43
+ """ # noqa: E501
44
+
45
+ async def determine_response(self, agent_message: str) -> ResponseType:
46
+ """
47
+ Determine whether the agent should stop or continue based on its message.
48
+
49
+ Args:
50
+ agent_message: The message from the agent
51
+
52
+ Returns:
53
+ ResponseType: Either "STOP" or "CONTINUE"
54
+ """
55
+ try:
56
+ response = await self.client.chat.completions.create(
57
+ model="gpt-4o",
58
+ messages=[
59
+ {"role": "system", "content": self.system_prompt},
60
+ {
61
+ "role": "user",
62
+ "content": f"Agent message: {agent_message}\n\nWhat is the appropriate response?", # noqa: E501
63
+ },
64
+ ],
65
+ temperature=0.1, # Low temperature for more deterministic responses
66
+ max_tokens=5, # We only need a short response
67
+ )
68
+
69
+ response_text = response.choices[0].message.content
70
+ if not response_text:
71
+ return "CONTINUE"
72
+
73
+ response_text = response_text.strip().upper()
74
+
75
+ # Validate the response
76
+ if "STOP" in response_text:
77
+ return "STOP"
78
+ else:
79
+ return "CONTINUE"
80
+
81
+ except Exception:
82
+ return "CONTINUE" # Default to continue on error