APEX/BrowserUse_and_ComputerUse_skills
5
0
Fork 0
Code Issues 6 Pull requests 1 Projects 1 Releases Packages Wiki Activity Actions

some cleanups

Browse source
This commit is contained in:
teknium 2025-11-05 03:47:17 +00:00
parent ab5c9fc37b
commit c82741c3d8
15 changed files with 911 additions and 56 deletions
Download patch file Download diff file Expand all files Collapse all files

4
tools/image_generation_tool.py
View file

@ -414,7 +414,7 @@ async def image_generate_tool(
_log_debug_call("image_generate_tool", debug_call_data)
_save_debug_log()
return json.dumps(response_data, indent=2)
return json.dumps(response_data, indent=2, ensure_ascii=False)
except Exception as e:
generation_time = (datetime.datetime.now() - start_time).total_seconds()
@ -432,7 +432,7 @@ async def image_generate_tool(
_log_debug_call("image_generate_tool", debug_call_data)
_save_debug_log()
return json.dumps(response_data, indent=2)
return json.dumps(response_data, indent=2, ensure_ascii=False)
def check_fal_api_key() -> bool:

4
tools/mixture_of_agents_tool.py
View file

@ -410,7 +410,7 @@ async def mixture_of_agents_tool(
_log_debug_call("mixture_of_agents_tool", debug_call_data)
_save_debug_log()
return json.dumps(result, indent=2)
return json.dumps(result, indent=2, ensure_ascii=False)
except Exception as e:
error_msg = f"Error in MoA processing: {str(e)}"
@ -436,7 +436,7 @@ async def mixture_of_agents_tool(
_log_debug_call("mixture_of_agents_tool", debug_call_data)
_save_debug_log()
return json.dumps(result, indent=2)
return json.dumps(result, indent=2, ensure_ascii=False)
def check_nous_api_key() -> bool:

8
tools/terminal_tool.py
View file

@ -139,7 +139,7 @@ def terminal_tool(
"exit_code": -1,
"error": f"Terminal tool is disabled due to import error: {import_error}",
"status": "disabled"
})
}, ensure_ascii=False)
# Get configuration from environment
vm_lifetime_seconds = int(os.getenv("HECATE_VM_LIFETIME_SECONDS", "300"))
@ -155,7 +155,7 @@ def terminal_tool(
"exit_code": -1,
"error": "MORPH_API_KEY environment variable not set",
"status": "disabled"
})
}, ensure_ascii=False)
# Get or create VM instance and execution context
# This is critical for interactive session support - the context must persist!
@ -219,7 +219,7 @@ def terminal_tool(
"status": "active" if result.get("session_id") else "ended"
}
return json.dumps(formatted_result)
return json.dumps(formatted_result, ensure_ascii=False)
except Exception as e:
return json.dumps({
@ -229,7 +229,7 @@ def terminal_tool(
"exit_code": -1,
"error": f"Failed to execute terminal command: {str(e)}",
"status": "error"
})
}, ensure_ascii=False)
def check_hecate_requirements() -> bool:
"""

4
tools/vision_tools.py
View file

@ -346,7 +346,7 @@ async def vision_analyze_tool(
_log_debug_call("vision_analyze_tool", debug_call_data)
_save_debug_log()
return json.dumps(result, indent=2)
return json.dumps(result, indent=2, ensure_ascii=False)
except Exception as e:
error_msg = f"Error analyzing image: {str(e)}"
@ -362,7 +362,7 @@ async def vision_analyze_tool(
_log_debug_call("vision_analyze_tool", debug_call_data)
_save_debug_log()
return json.dumps(result, indent=2)
return json.dumps(result, indent=2, ensure_ascii=False)
finally:
# Clean up temporary image file

69
tools/web_tools.py
View file

@ -183,16 +183,33 @@ Your goal is to preserve ALL important information while reducing length. Never
Create a markdown summary that captures all key information in a well-organized, scannable format. Include important quotes and code snippets in their original formatting. Focus on actionable information, specific details, and unique insights."""
# Call the LLM asynchronously
response = await nous_client.chat.completions.create(
model=model,
messages=[
{"role": "system", "content": system_prompt},
{"role": "user", "content": user_prompt}
],
temperature=0.1, # Low temperature for consistent extraction
max_tokens=4000 # Generous limit for comprehensive processing
)
# Call the LLM asynchronously with retry logic for flaky API
max_retries = 3
retry_delay = 2 # Start with 2 seconds
last_error = None
for attempt in range(max_retries):
try:
response = await nous_client.chat.completions.create(
model=model,
messages=[
{"role": "system", "content": system_prompt},
{"role": "user", "content": user_prompt}
],
temperature=0.1, # Low temperature for consistent extraction
max_tokens=4000 # Generous limit for comprehensive processing
)
break # Success, exit retry loop
except Exception as api_error:
last_error = api_error
if attempt < max_retries - 1:
print(f"⚠️ LLM API call failed (attempt {attempt + 1}/{max_retries}): {str(api_error)[:100]}")
print(f" Retrying in {retry_delay}s...")
await asyncio.sleep(retry_delay)
retry_delay *= 2 # Exponential backoff: 2s, 4s, 8s
else:
# All retries exhausted
raise last_error
# Get the markdown response directly
processed_content = response.choices[0].message.content.strip()
@ -344,7 +361,7 @@ def web_search_tool(query: str, limit: int = 5) -> str:
debug_call_data["results_count"] = results_count
# Convert to JSON
result_json = json.dumps(response_data, indent=2)
result_json = json.dumps(response_data, indent=2, ensure_ascii=False)
debug_call_data["final_response_size"] = len(result_json)
@ -362,7 +379,7 @@ def web_search_tool(query: str, limit: int = 5) -> str:
_log_debug_call("web_search_tool", debug_call_data)
_save_debug_log()
return json.dumps({"error": error_msg})
return json.dumps({"error": error_msg}, ensure_ascii=False)
async def web_extract_tool(
@ -575,18 +592,20 @@ async def web_extract_tool(
"title": r.get("title", ""),
"content": r.get("content", ""),
"error": r.get("error"),
**({"llm_model": model} if use_llm_processing else {})
}
for r in response.get("results", [])
]
trimmed_response = {"results": trimmed_results}
# Include model name used for summarization when LLM processing was requested
if use_llm_processing:
trimmed_response["llm_model"] = model
if trimmed_response.get("results") == []:
result_json = json.dumps({"error": "Content was inaccessible or not found"}, ensure_ascii=False)
cleaned_result = clean_base64_images(result_json)
result_json = json.dumps(trimmed_response, indent=2)
# Clean base64 images from extracted content
cleaned_result = clean_base64_images(result_json)
else:
result_json = json.dumps(trimmed_response, indent=2, ensure_ascii=False)
cleaned_result = clean_base64_images(result_json)
debug_call_data["final_response_size"] = len(cleaned_result)
debug_call_data["processing_applied"].append("base64_image_removal")
@ -605,7 +624,7 @@ async def web_extract_tool(
_log_debug_call("web_extract_tool", debug_call_data)
_save_debug_log()
return json.dumps({"error": error_msg})
return json.dumps({"error": error_msg}, ensure_ascii=False)
async def web_crawl_tool(
@ -851,17 +870,13 @@ async def web_crawl_tool(
{
"title": r.get("title", ""),
"content": r.get("content", ""),
"error": r.get("error"),
**({"llm_model": model} if use_llm_processing else {})
"error": r.get("error")
}
for r in response.get("results", [])
]
trimmed_response = {"results": trimmed_results}
# Include model name used for summarization when LLM processing was requested
if use_llm_processing:
trimmed_response["llm_model"] = model
result_json = json.dumps(trimmed_response, indent=2)
result_json = json.dumps(trimmed_response, indent=2, ensure_ascii=False)
# Clean base64 images from crawled content
cleaned_result = clean_base64_images(result_json)
@ -882,7 +897,7 @@ async def web_crawl_tool(
_log_debug_call("web_crawl_tool", debug_call_data)
_save_debug_log()
return json.dumps({"error": error_msg})
return json.dumps({"error": error_msg}, ensure_ascii=False)
# Convenience function to check if API key is available