Cleanup time!

batch_runner.py

@@ -27,7 +27,7 @@ import time
 from pathlib import Path
 from typing import List, Dict, Any, Optional, Tuple
 from datetime import datetime
-from multiprocessing import Pool, Manager, Lock
+from multiprocessing import Pool, Lock
 import traceback
 
 from rich.progress import Progress, SpinnerColumn, BarColumn, TextColumn, TimeRemainingColumn, MofNCompleteColumn
@@ -36,7 +36,6 @@ import fire
 
 from run_agent import AIAgent
 from toolset_distributions import (
-    get_distribution, 
     list_distributions, 
     sample_toolsets_from_distribution,
     validate_distribution
@@ -173,7 +172,7 @@ def _extract_tool_stats(messages: List[Dict[str, Any]]) -> Dict[str, Dict[str, i
                     if content_json.get("success") is False:
                         is_success = False
                         
-            except:
+            except (json.JSONDecodeError, ValueError, TypeError):
                 # If not JSON, check if content is empty or explicitly states an error
                 # Note: We avoid simple substring matching to prevent false positives
                 if not content:

cli.py

@@ -39,16 +39,16 @@ from prompt_toolkit.layout.menus import CompletionsMenu
 from prompt_toolkit.widgets import TextArea
 from prompt_toolkit.key_binding import KeyBindings
 from prompt_toolkit.completion import Completer, Completion
-from prompt_toolkit.keys import Keys
 from prompt_toolkit import print_formatted_text as _pt_print
 from prompt_toolkit.formatted_text import ANSI as _PT_ANSI
 import threading
 import queue
-import tempfile
 
 
 # Load environment variables first
 from dotenv import load_dotenv
+from hermes_constants import OPENROUTER_BASE_URL
+
 env_path = Path(__file__).parent / '.env'
 if env_path.exists():
     load_dotenv(dotenv_path=env_path)
@@ -88,7 +88,7 @@ def load_cli_config() -> Dict[str, Any]:
     defaults = {
         "model": {
             "default": "anthropic/claude-opus-4.6",
-            "base_url": "https://openrouter.ai/api/v1",
+            "base_url": OPENROUTER_BASE_URL,
             "provider": "auto",
         },
         "terminal": {
@@ -262,20 +262,15 @@ def load_cli_config() -> Dict[str, Any]:
 # Load configuration at module startup
 CLI_CONFIG = load_cli_config()
 
-from rich.console import Console, Group
+from rich.console import Console
 from rich.panel import Panel
-from rich.text import Text
 from rich.table import Table
-from rich.markdown import Markdown
-from rich.columns import Columns
-from rich.align import Align
-from rich import box
 
 import fire
 
 # Import the agent and tool systems
 from run_agent import AIAgent
-from model_tools import get_tool_definitions, get_all_tool_names, get_toolset_for_tool, get_available_toolsets
+from model_tools import get_tool_definitions, get_toolset_for_tool
 from toolsets import get_all_toolsets, get_toolset_info, resolve_toolset, validate_toolset
 
 # Cron job system for scheduled tasks

gateway/delivery.py

@@ -8,14 +8,13 @@ Routes messages to the appropriate destination based on:
 - Local (always saved to files)
 """
 
-import json
 from pathlib import Path
 from datetime import datetime
 from dataclasses import dataclass
 from typing import Dict, List, Optional, Any, Union
 from enum import Enum
 
-from .config import Platform, GatewayConfig, HomeChannel
+from .config import Platform, GatewayConfig
 from .session import SessionSource
 
 

gateway/run.py

@@ -1081,7 +1081,7 @@ class GatewayRunner:
                         try:
                             msg = progress_queue.get_nowait()
                             await adapter.send(chat_id=source.chat_id, content=msg)
-                        except:
+                        except Exception:
                             break
                     return
                 except Exception as e:

hermes_cli/auth.py

@@ -30,6 +30,7 @@ import httpx
 import yaml
 
 from hermes_cli.config import get_hermes_home, get_config_path
+from hermes_constants import OPENROUTER_BASE_URL
 
 try:
     import fcntl
@@ -865,7 +866,7 @@ def _reset_config_provider() -> Path:
     if isinstance(model, dict):
         model["provider"] = "auto"
         if "base_url" in model:
-            model["base_url"] = "https://openrouter.ai/api/v1"
+            model["base_url"] = OPENROUTER_BASE_URL
     config_path.write_text(yaml.safe_dump(config, sort_keys=False))
     return config_path
 

hermes_cli/colors.py

@@ -0,0 +1,22 @@
+"""Shared ANSI color utilities for Hermes CLI modules."""
+
+import sys
+
+
+class Colors:
+    RESET = "\033[0m"
+    BOLD = "\033[1m"
+    DIM = "\033[2m"
+    RED = "\033[31m"
+    GREEN = "\033[32m"
+    YELLOW = "\033[33m"
+    BLUE = "\033[34m"
+    MAGENTA = "\033[35m"
+    CYAN = "\033[36m"
+
+
+def color(text: str, *codes) -> str:
+    """Apply color codes to text (only when output is a TTY)."""
+    if not sys.stdout.isatty():
+        return text
+    return "".join(codes) + text + Colors.RESET

hermes_cli/config.py

@@ -20,22 +20,7 @@ from typing import Dict, Any, Optional, List, Tuple
 
 import yaml
 
-# ANSI colors
+from hermes_cli.colors import Colors, color
-class Colors:
-    RESET = "\033[0m"
-    BOLD = "\033[1m"
-    DIM = "\033[2m"
-    RED = "\033[31m"
-    GREEN = "\033[32m"
-    YELLOW = "\033[33m"
-    BLUE = "\033[34m"
-    MAGENTA = "\033[35m"
-    CYAN = "\033[36m"
-
-def color(text: str, *codes) -> str:
-    if not sys.stdout.isatty():
-        return text
-    return "".join(codes) + text + Colors.RESET
 
 
 # =============================================================================

hermes_cli/cron.py

@@ -4,29 +4,14 @@ Cron subcommand for hermes CLI.
 Handles: hermes cron [list|daemon|tick]
 """
 
-import json
 import sys
-import time
 from pathlib import Path
 from datetime import datetime
 
 PROJECT_ROOT = Path(__file__).parent.parent.resolve()
 sys.path.insert(0, str(PROJECT_ROOT))
 
-# ANSI colors
+from hermes_cli.colors import Colors, color
-class Colors:
-    RESET = "\033[0m"
-    BOLD = "\033[1m"
-    DIM = "\033[2m"
-    RED = "\033[31m"
-    GREEN = "\033[32m"
-    YELLOW = "\033[33m"
-    CYAN = "\033[36m"
-
-def color(text: str, *codes) -> str:
-    if not sys.stdout.isatty():
-        return text
-    return "".join(codes) + text + Colors.RESET
 
 
 def cron_list(show_all: bool = False):

hermes_cli/doctor.py

@@ -23,20 +23,8 @@ if _env_path.exists():
 # Also try project .env as fallback
 load_dotenv(PROJECT_ROOT / ".env", override=False)
 
-# ANSI colors
+from hermes_cli.colors import Colors, color
-class Colors:
+from hermes_constants import OPENROUTER_MODELS_URL
-    RESET = "\033[0m"
-    BOLD = "\033[1m"
-    DIM = "\033[2m"
-    RED = "\033[31m"
-    GREEN = "\033[32m"
-    YELLOW = "\033[33m"
-    CYAN = "\033[36m"
-
-def color(text: str, *codes) -> str:
-    if not sys.stdout.isatty():
-        return text
-    return "".join(codes) + text + Colors.RESET
 
 def check_ok(text: str, detail: str = ""):
     print(f"  {color('✓', Colors.GREEN)} {text}" + (f" {color(detail, Colors.DIM)}" if detail else ""))
@@ -314,7 +302,7 @@ def run_doctor(args):
         try:
             import httpx
             response = httpx.get(
-                "https://openrouter.ai/api/v1/models",
+                OPENROUTER_MODELS_URL,
                 headers={"Authorization": f"Bearer {openrouter_key}"},
                 timeout=10
             )

hermes_cli/main.py

@@ -40,6 +40,7 @@ if env_path.exists():
     load_dotenv(dotenv_path=env_path)
 
 from hermes_cli import __version__
+from hermes_constants import OPENROUTER_BASE_URL
 
 
 def cmd_chat(args):
@@ -241,7 +242,7 @@ def _model_flow_openrouter(config, current_model=""):
         model = cfg.get("model")
         if isinstance(model, dict):
             model["provider"] = "openrouter"
-            model["base_url"] = "https://openrouter.ai/api/v1"
+            model["base_url"] = OPENROUTER_BASE_URL
         save_config(cfg)
         deactivate_provider()
         print(f"Default model set to: {selected} (via OpenRouter)")

hermes_cli/pairing.py

@@ -8,9 +8,6 @@ Usage:
     hermes pairing clear-pending     # Clear all expired/pending codes
 """
 
-import sys
-
-
 def pairing_command(args):
     """Handle hermes pairing subcommands."""
     from gateway.pairing import PairingStore

hermes_cli/setup.py

@@ -26,23 +26,7 @@ from hermes_cli.config import (
     ensure_hermes_home, DEFAULT_CONFIG
 )
 
-# ANSI colors
+from hermes_cli.colors import Colors, color
-class Colors:
-    RESET = "\033[0m"
-    BOLD = "\033[1m"
-    DIM = "\033[2m"
-    RED = "\033[31m"
-    GREEN = "\033[32m"
-    YELLOW = "\033[33m"
-    BLUE = "\033[34m"
-    MAGENTA = "\033[35m"
-    CYAN = "\033[36m"
-
-def color(text: str, *codes) -> str:
-    """Apply color codes to text."""
-    if not sys.stdout.isatty():
-        return text
-    return "".join(codes) + text + Colors.RESET
 
 def print_header(title: str):
     """Print a section header."""

hermes_cli/skills_hub.py

@@ -12,7 +12,6 @@ handler are thin wrappers that parse args and delegate.
 
 import json
 import shutil
-import sys
 from pathlib import Path
 from typing import Optional
 

hermes_cli/status.py

@@ -11,20 +11,8 @@ from pathlib import Path
 
 PROJECT_ROOT = Path(__file__).parent.parent.resolve()
 
-# ANSI colors
+from hermes_cli.colors import Colors, color
-class Colors:
+from hermes_constants import OPENROUTER_MODELS_URL
-    RESET = "\033[0m"
-    BOLD = "\033[1m"
-    DIM = "\033[2m"
-    RED = "\033[31m"
-    GREEN = "\033[32m"
-    YELLOW = "\033[33m"
-    CYAN = "\033[36m"
-
-def color(text: str, *codes) -> str:
-    if not sys.stdout.isatty():
-        return text
-    return "".join(codes) + text + Colors.RESET
 
 def check_mark(ok: bool) -> str:
     if ok:
@@ -232,7 +220,7 @@ def show_status(args):
                 jobs = data.get("jobs", [])
                 enabled_jobs = [j for j in jobs if j.get("enabled", True)]
                 print(f"  Jobs:         {len(enabled_jobs)} active, {len(jobs)} total")
-        except:
+        except Exception:
             print(f"  Jobs:         (error reading jobs file)")
     else:
         print(f"  Jobs:         0")
@@ -250,7 +238,7 @@ def show_status(args):
             with open(sessions_file) as f:
                 data = json.load(f)
                 print(f"  Active:       {len(data)} session(s)")
-        except:
+        except Exception:
             print(f"  Active:       (error reading sessions file)")
     else:
         print(f"  Active:       0")
@@ -268,7 +256,7 @@ def show_status(args):
             try:
                 import httpx
                 response = httpx.get(
-                    "https://openrouter.ai/api/v1/models",
+                    OPENROUTER_MODELS_URL,
                     headers={"Authorization": f"Bearer {openrouter_key}"},
                     timeout=10
                 )
@@ -288,7 +276,7 @@ def show_status(args):
             port_in_use = result == 0
             # This is informational, not necessarily bad
             print(f"  Port 18789:   {'in use' if port_in_use else 'available'}")
-        except:
+        except OSError:
             pass
     
     print()

hermes_cli/uninstall.py

@@ -13,23 +13,7 @@ import subprocess
 from pathlib import Path
 from typing import Optional
 
-# ANSI colors
+from hermes_cli.colors import Colors, color
-class Colors:
-    RESET = "\033[0m"
-    BOLD = "\033[1m"
-    DIM = "\033[2m"
-    RED = "\033[31m"
-    GREEN = "\033[32m"
-    YELLOW = "\033[33m"
-    BLUE = "\033[34m"
-    MAGENTA = "\033[35m"
-    CYAN = "\033[36m"
-
-def color(text: str, *codes) -> str:
-    """Apply color codes to text (only in TTY)."""
-    if not sys.stdout.isatty():
-        return text
-    return "".join(codes) + text + Colors.RESET
 
 def log_info(msg: str):
     print(f"{color('→', Colors.CYAN)} {msg}")

hermes_constants.py

@@ -0,0 +1,9 @@
+"""Shared constants for Hermes Agent.
+
+Import-safe module with no dependencies — can be imported from anywhere
+without risk of circular imports.
+"""
+
+OPENROUTER_BASE_URL = "https://openrouter.ai/api/v1"
+OPENROUTER_MODELS_URL = f"{OPENROUTER_BASE_URL}/models"
+OPENROUTER_CHAT_URL = f"{OPENROUTER_BASE_URL}/chat/completions"

model_tools.py

@@ -31,19 +31,17 @@ import asyncio
 import os
 from typing import Dict, Any, List, Optional, Tuple
 
-from tools.web_tools import web_search_tool, web_extract_tool, web_crawl_tool, check_firecrawl_api_key
+from tools.web_tools import web_search_tool, web_extract_tool, check_firecrawl_api_key
-from tools.terminal_tool import terminal_tool, check_terminal_requirements, TERMINAL_TOOL_DESCRIPTION, cleanup_vm
+from tools.terminal_tool import terminal_tool, check_terminal_requirements, TERMINAL_TOOL_DESCRIPTION
 # File manipulation tools (read, write, patch, search)
 from tools.file_tools import read_file_tool, write_file_tool, patch_tool, search_tool
 from tools import check_file_requirements
-# Hecate/MorphCloud terminal tool (cloud VMs) - available as alternative backend
-from tools.terminal_hecate import terminal_hecate_tool, check_hecate_requirements, TERMINAL_HECATE_DESCRIPTION
 from tools.vision_tools import vision_analyze_tool, check_vision_requirements
 from tools.mixture_of_agents_tool import mixture_of_agents_tool, check_moa_requirements
 from tools.image_generation_tool import image_generate_tool, check_image_generation_requirements
-from tools.skills_tool import skills_list, skill_view, check_skills_requirements, SKILLS_TOOL_DESCRIPTION
+from tools.skills_tool import skills_list, skill_view, check_skills_requirements
 # Agent-managed skill creation/editing
-from tools.skill_manager_tool import skill_manage, check_skill_manage_requirements, SKILL_MANAGE_SCHEMA
+from tools.skill_manager_tool import skill_manage, SKILL_MANAGE_SCHEMA
 # RL Training tools (Tinker-Atropos)
 from tools.rl_training_tool import (
     rl_list_environments,
@@ -64,7 +62,6 @@ from tools.cronjob_tools import (
     list_cronjobs,
     remove_cronjob,
     check_cronjob_requirements,
-    get_cronjob_tool_definitions,
     SCHEDULE_CRONJOB_SCHEMA,
     LIST_CRONJOBS_SCHEMA,
     REMOVE_CRONJOB_SCHEMA
@@ -99,11 +96,7 @@ from tools.clarify_tool import clarify_tool, check_clarify_requirements, CLARIFY
 from tools.code_execution_tool import execute_code, check_sandbox_requirements, EXECUTE_CODE_SCHEMA
 # Subagent delegation
 from tools.delegate_tool import delegate_task, check_delegate_requirements, DELEGATE_TASK_SCHEMA
-from toolsets import (
+from toolsets import resolve_toolset, validate_toolset
-    get_toolset, resolve_toolset, resolve_multiple_toolsets,
-    get_all_toolsets, get_toolset_names, validate_toolset,
-    get_toolset_info, print_toolset_tree
-)
 
 
 # =============================================================================
@@ -262,43 +255,6 @@ def check_tool_availability(quiet: bool = False) -> Tuple[List[str], List[Dict[s
     return available, unavailable
 
 
-def print_tool_availability_warnings(unavailable: List[Dict[str, Any]], prefix: str = ""):
-    """Print warnings about unavailable tools."""
-    if not unavailable:
-        return
-    
-    # Filter to only those missing API keys (not system dependencies)
-    api_key_missing = [u for u in unavailable if u["missing_vars"]]
-    
-    if api_key_missing:
-        print(f"{prefix}⚠️  Some tools are disabled due to missing API keys:")
-        for item in api_key_missing:
-            vars_str = ", ".join(item["missing_vars"])
-            print(f"{prefix}   • {item['name']}: missing {vars_str}")
-            if item["setup_url"]:
-                print(f"{prefix}     Get key at: {item['setup_url']}")
-        print(f"{prefix}   Run 'hermes setup' to configure API keys")
-        print()
-
-
-def get_tool_availability_summary() -> Dict[str, Any]:
-    """
-    Get a summary of tool availability for display in status/doctor commands.
-    
-    Returns:
-        Dict with 'available' and 'unavailable' lists of tool info
-    """
-    available, unavailable = check_tool_availability()
-    
-    return {
-        "available": [
-            {"id": tid, "name": TOOLSET_REQUIREMENTS[tid]["name"], "tools": TOOLSET_REQUIREMENTS[tid]["tools"]}
-            for tid in available
-        ],
-        "unavailable": unavailable,
-    }
-
-
 def get_web_tool_definitions() -> List[Dict[str, Any]]:
     """
     Get tool definitions for web tools in OpenAI's expected format.

pyproject.toml

@@ -62,7 +62,7 @@ hermes = "hermes_cli.main:main"
 hermes-agent = "run_agent:main"
 
 [tool.setuptools]
-py-modules = ["run_agent", "model_tools", "toolsets", "batch_runner", "trajectory_compressor", "toolset_distributions", "cli"]
+py-modules = ["run_agent", "model_tools", "toolsets", "batch_runner", "trajectory_compressor", "toolset_distributions", "cli", "hermes_constants"]
 
 [tool.setuptools.packages.find]
 include = ["tools", "hermes_cli", "gateway", "cron"]

requirements.txt

@@ -1,3 +1,7 @@
+# NOTE: This file is maintained for convenience only.
+# The canonical dependency list is in pyproject.toml.
+# Preferred install: pip install -e ".[all]"
+
 # Core dependencies
 openai
 python-dotenv
@@ -10,6 +14,7 @@ pyyaml
 requests
 jinja2
 pydantic>=2.0
+PyJWT[crypto]
 
 # Web tools
 firecrawl-py
@@ -23,27 +28,13 @@ litellm>=1.75.5
 typer
 platformdirs
 
-# Optional: For Docker backend (recommended)
-# Requires Docker installed and user in 'docker' group
-
-# Optional: For Modal backend (cloud execution)
-# swe-rex[modal]>=1.4.0  # Includes modal + boto3 + swe-rex runtime
-
 # Text-to-speech (Edge TTS is free, no API key needed)
 edge-tts
 
-# Optional: Premium TTS providers
-# elevenlabs  # Uncomment if using ElevenLabs TTS (needs ELEVENLABS_API_KEY)
-
 # Optional: For cron expression parsing (cronjob scheduling)
 croniter
 
 # Optional: For messaging platform integrations (gateway)
-# Telegram
 python-telegram-bot>=20.0
-
-# Discord
 discord.py>=2.0
-
-# WhatsApp bridge communication + general async HTTP (used by gateway)
 aiohttp>=3.9.0

rl_cli.py

@@ -64,8 +64,10 @@ from tools.rl_training_tool import check_rl_api_keys, get_missing_keys
 # Config Loading
 # ============================================================================
 
+from hermes_constants import OPENROUTER_BASE_URL
+
 DEFAULT_MODEL = "anthropic/claude-opus-4.5"
-DEFAULT_BASE_URL = "https://openrouter.ai/api/v1"
+DEFAULT_BASE_URL = OPENROUTER_BASE_URL
 
 
 def load_hermes_config() -> dict:

run_agent.py

@@ -25,6 +25,7 @@ import json
 import logging
 import os
 import random
+import re
 import sys
 import time
 import threading
@@ -54,6 +55,8 @@ from tools.browser_tool import cleanup_browser
 
 import requests
 
+from hermes_constants import OPENROUTER_BASE_URL, OPENROUTER_MODELS_URL
+
 # =============================================================================
 # Default Agent Identity & Platform Hints
 # =============================================================================
@@ -133,7 +136,7 @@ def fetch_model_metadata(force_refresh: bool = False) -> Dict[str, Dict[str, Any
     
     try:
         response = requests.get(
-            "https://openrouter.ai/api/v1/models",
+            OPENROUTER_MODELS_URL,
             timeout=10
         )
         response.raise_for_status()
@@ -282,7 +285,7 @@ class ContextCompressor:
         api_key = os.getenv("OPENROUTER_API_KEY", "")
         self.client = OpenAI(
             api_key=api_key,
-            base_url="https://openrouter.ai/api/v1"
+            base_url=OPENROUTER_BASE_URL
         ) if api_key else None
     
     def update_from_response(self, usage: Dict[str, Any]):
@@ -600,7 +603,6 @@ def build_skills_system_prompt() -> str:
         str: The skills system prompt section, or empty string if no skills found.
     """
     import os
-    import re
     from pathlib import Path
     
     hermes_home = Path(os.getenv("HERMES_HOME", Path.home() / ".hermes"))
@@ -1093,8 +1095,8 @@ class AIAgent:
         Args:
             base_url (str): Base URL for the model API (optional)
             api_key (str): API key for authentication (optional, uses env var if not provided)
-            model (str): Model name to use (default: "gpt-4")
+            model (str): Model name to use (default: "anthropic/claude-opus-4.6")
-            max_iterations (int): Maximum number of tool calling iterations (default: 10)
+            max_iterations (int): Maximum number of tool calling iterations (default: 60)
             tool_delay (float): Delay between tool calls in seconds (default: 1.0)
             enabled_toolsets (List[str]): Only enable tools from these toolsets (optional)
             disabled_toolsets (List[str]): Disable tools from these toolsets (optional)
@@ -1102,7 +1104,7 @@ class AIAgent:
             verbose_logging (bool): Enable verbose logging for debugging (default: False)
             quiet_mode (bool): Suppress progress output for clean CLI experience (default: False)
             ephemeral_system_prompt (str): System prompt used during agent execution but NOT saved to trajectories (optional)
-            log_prefix_chars (int): Number of characters to show in log previews for tool calls/responses (default: 20)
+            log_prefix_chars (int): Number of characters to show in log previews for tool calls/responses (default: 100)
             log_prefix (str): Prefix to add to all log messages for identification in parallel processing (default: "")
             providers_allowed (List[str]): OpenRouter providers to allow (optional)
             providers_ignored (List[str]): OpenRouter providers to ignore (optional)
@@ -1137,7 +1139,7 @@ class AIAgent:
         self.log_prefix = f"{log_prefix} " if log_prefix else ""
         # Store effective base URL for feature detection (prompt caching, reasoning, etc.)
         # When no base_url is provided, the client defaults to OpenRouter, so reflect that here.
-        self.base_url = base_url or "https://openrouter.ai/api/v1"
+        self.base_url = base_url or OPENROUTER_BASE_URL
         self.tool_progress_callback = tool_progress_callback
         self.clarify_callback = clarify_callback
         self._last_reported_tool = None  # Track for "new tool" mode
@@ -1215,7 +1217,7 @@ class AIAgent:
         if base_url:
             client_kwargs["base_url"] = base_url
         else:
-            client_kwargs["base_url"] = "https://openrouter.ai/api/v1"
+            client_kwargs["base_url"] = OPENROUTER_BASE_URL
         
         # Handle API key - OpenRouter is the primary provider
         if api_key:
@@ -1636,7 +1638,6 @@ class AIAgent:
         if not content:
             return False
         
-        import re
         # Remove all <think>...</think> blocks (including nested ones, non-greedy)
         cleaned = re.sub(r'<think>.*?</think>', '', content, flags=re.DOTALL)
         
@@ -1686,6 +1687,19 @@ class AIAgent:
         
         return None
     
+    def _cleanup_task_resources(self, task_id: str) -> None:
+        """Clean up VM and browser resources for a given task."""
+        try:
+            cleanup_vm(task_id)
+        except Exception as e:
+            if self.verbose_logging:
+                logging.warning(f"Failed to cleanup VM for task {task_id}: {e}")
+        try:
+            cleanup_browser(task_id)
+        except Exception as e:
+            if self.verbose_logging:
+                logging.warning(f"Failed to cleanup browser for task {task_id}: {e}")
+
     def _get_messages_up_to_last_assistant(self, messages: List[Dict]) -> List[Dict]:
         """
         Get messages up to (but not including) the last assistant turn.
@@ -2331,7 +2345,6 @@ class AIAgent:
             Dict: Complete conversation result with final response and message history
         """
         # Generate unique task_id if not provided to isolate VMs between concurrent tasks
-        import uuid
         effective_task_id = task_id or str(uuid.uuid4())
         
         # Reset retry counters at the start of each conversation to prevent state leakage
@@ -2628,17 +2641,7 @@ class AIAgent:
                             print(f"{self.log_prefix}   ⏪ Rolling back to last complete assistant turn")
                             rolled_back_messages = self._get_messages_up_to_last_assistant(messages)
                             
-                            # Clean up VM and browser
+                            self._cleanup_task_resources(effective_task_id)
-                            try:
-                                cleanup_vm(effective_task_id)
-                            except Exception as e:
-                                if self.verbose_logging:
-                                    logging.warning(f"Failed to cleanup VM for task {effective_task_id}: {e}")
-                            try:
-                                cleanup_browser(effective_task_id)
-                            except Exception as e:
-                                if self.verbose_logging:
-                                    logging.warning(f"Failed to cleanup browser for task {effective_task_id}: {e}")
                             
                             return {
                                 "final_response": None,
@@ -2846,15 +2849,7 @@ class AIAgent:
                         self._incomplete_scratchpad_retries = 0
                         
                         rolled_back_messages = self._get_messages_up_to_last_assistant(messages)
-                        
+                        self._cleanup_task_resources(effective_task_id)
-                        try:
-                            cleanup_vm(effective_task_id)
-                        except Exception:
-                            pass
-                        try:
-                            cleanup_browser(effective_task_id)
-                        except Exception:
-                            pass
                         
                         return {
                             "final_response": None,
@@ -3247,18 +3242,7 @@ class AIAgent:
                             self._empty_content_retries = 0  # Reset for next conversation
                             
                             rolled_back_messages = self._get_messages_up_to_last_assistant(messages)
-                            
+                            self._cleanup_task_resources(effective_task_id)
-                            # Clean up VM and browser
-                            try:
-                                cleanup_vm(effective_task_id)
-                            except Exception as e:
-                                if self.verbose_logging:
-                                    logging.warning(f"Failed to cleanup VM for task {effective_task_id}: {e}")
-                            try:
-                                cleanup_browser(effective_task_id)
-                            except Exception as e:
-                                if self.verbose_logging:
-                                    logging.warning(f"Failed to cleanup browser for task {effective_task_id}: {e}")
                             
                             return {
                                 "final_response": None,
@@ -3365,7 +3349,6 @@ class AIAgent:
                     final_response = summary_response.choices[0].message.content
                     # Strip think blocks from final response
                     if "<think>" in final_response:
-                        import re
                         final_response = re.sub(r'<think>.*?</think>\s*', '', final_response, flags=re.DOTALL).strip()
                     
                     # Add to messages for session continuity
@@ -3384,17 +3367,7 @@ class AIAgent:
         self._save_trajectory(messages, user_message, completed)
 
         # Clean up VM and browser for this task after conversation completes
-        try:
+        self._cleanup_task_resources(effective_task_id)
-            cleanup_vm(effective_task_id)
-        except Exception as e:
-            if self.verbose_logging:
-                logging.warning(f"Failed to cleanup VM for task {effective_task_id}: {e}")
-        
-        try:
-            cleanup_browser(effective_task_id)
-        except Exception as e:
-            if self.verbose_logging:
-                logging.warning(f"Failed to cleanup browser for task {effective_task_id}: {e}")
 
         # Update session messages and save session log
         self._session_messages = messages
@@ -3644,7 +3617,6 @@ def main(
     
     # Save sample trajectory to UUID-named file if requested
     if save_sample:
-        import uuid
         sample_id = str(uuid.uuid4())[:8]
         sample_filename = f"sample_{sample_id}.json"
         

scripts/sample_and_compress.py

@@ -108,7 +108,7 @@ def _count_tokens_for_entry(entry: Dict) -> Tuple[Dict, int]:
         if value:
             try:
                 total += len(_TOKENIZER.encode(value))
-            except:
+            except Exception:
                 # Fallback to character estimate
                 total += len(value) // 4
     

tests/test_checkpoint_resumption.py

@@ -23,7 +23,6 @@ import os
 import shutil
 import sys
 import time
-import signal
 from pathlib import Path
 from typing import List, Dict, Any
 import traceback

tests/test_delegate.py

@@ -12,7 +12,6 @@ Run with:  python -m pytest tests/test_delegate.py -v
 import json
 import os
 import sys
-import time
 import unittest
 from unittest.mock import MagicMock, patch
 

tests/test_temperature_fix.py

@@ -84,7 +84,7 @@ Create a markdown summary that captures all key information in a well-organized,
             max_tokens=4000
         )
         print(f"✅ SUCCESS")
-    except:
+    except Exception:
         print(f"❌ FAILED")
     
     await asyncio.sleep(0.5)
@@ -101,7 +101,7 @@ Create a markdown summary that captures all key information in a well-organized,
             max_tokens=4000
         )
         print(f"✅ SUCCESS")
-    except:
+    except Exception:
         print(f"❌ FAILED")
 
 if __name__ == "__main__":

tests/test_web_tools.py

@@ -21,7 +21,7 @@ import sys
 import os
 import argparse
 from datetime import datetime
-from typing import List, Dict, Any
+from typing import List
 
 # Import the web tools to test (updated path after moving tools/)
 from tools.web_tools import (

tools/browser_tool.py

@@ -57,6 +57,7 @@ import time
 import requests
 from typing import Dict, Any, Optional, List
 from pathlib import Path
+from hermes_constants import OPENROUTER_CHAT_URL
 
 # Try to import httpx for async LLM calls
 try:
@@ -821,7 +822,7 @@ Provide a concise summary focused on interactive elements and key content."""
     try:
         async with httpx.AsyncClient(timeout=30.0) as client:
             response = await client.post(
-                "https://openrouter.ai/api/v1/chat/completions",
+                OPENROUTER_CHAT_URL,
                 headers={
                     "Authorization": f"Bearer {api_key}",
                     "Content-Type": "application/json"
@@ -1324,7 +1325,7 @@ Focus on answering the user's specific question."""
             async def analyze_screenshot():
                 async with httpx.AsyncClient(timeout=60.0) as client:
                     response = await client.post(
-                        "https://openrouter.ai/api/v1/chat/completions",
+                        OPENROUTER_CHAT_URL,
                         headers={
                             "Authorization": f"Bearer {api_key}",
                             "Content-Type": "application/json"
@@ -1374,7 +1375,7 @@ Focus on answering the user's specific question."""
         else:
             # Fallback: use synchronous requests
             response = requests.post(
-                "https://openrouter.ai/api/v1/chat/completions",
+                OPENROUTER_CHAT_URL,
                 headers={
                     "Authorization": f"Bearer {api_key}",
                     "Content-Type": "application/json"

tools/code_execution_tool.py

@@ -345,7 +345,7 @@ def execute_code(
 
     # --- Set up temp directory with hermes_tools.py and script.py ---
     tmpdir = tempfile.mkdtemp(prefix="hermes_sandbox_")
-    sock_path = f"/tmp/hermes_rpc_{uuid.uuid4().hex}.sock"
+    sock_path = os.path.join(tempfile.gettempdir(), f"hermes_rpc_{uuid.uuid4().hex}.sock")
 
     tool_call_log: list = []
     tool_call_counter = [0]  # mutable so the RPC thread can increment

tools/fuzzy_match.py

@@ -446,33 +446,3 @@ def _map_normalized_positions(original: str, normalized: str,
         original_matches.append((orig_start, min(orig_end, len(original))))
     
     return original_matches
-
-
-# =============================================================================
-# Utility Functions
-# =============================================================================
-
-def find_best_match(content: str, pattern: str) -> Optional[Tuple[int, int, str]]:
-    """
-    Find the best match for a pattern and return the strategy name.
-    
-    Returns:
-        Tuple of (start, end, strategy_name) or None if no match
-    """
-    strategies = [
-        ("exact", _strategy_exact),
-        ("line_trimmed", _strategy_line_trimmed),
-        ("whitespace_normalized", _strategy_whitespace_normalized),
-        ("indentation_flexible", _strategy_indentation_flexible),
-        ("escape_normalized", _strategy_escape_normalized),
-        ("trimmed_boundary", _strategy_trimmed_boundary),
-        ("block_anchor", _strategy_block_anchor),
-        ("context_aware", _strategy_context_aware),
-    ]
-    
-    for strategy_name, strategy_fn in strategies:
-        matches = strategy_fn(content, pattern)
-        if matches:
-            return (matches[0][0], matches[0][1], strategy_name)
-    
-    return None

tools/mixture_of_agents_tool.py

@@ -53,8 +53,8 @@ import datetime
 from pathlib import Path
 from typing import Dict, Any, List, Optional
 from openai import AsyncOpenAI
+from hermes_constants import OPENROUTER_BASE_URL
 
-# Initialize OpenRouter API client lazily (only when needed)
 _openrouter_client = None
 
 def _get_openrouter_client():
@@ -66,7 +66,7 @@ def _get_openrouter_client():
             raise ValueError("OPENROUTER_API_KEY environment variable not set")
         _openrouter_client = AsyncOpenAI(
             api_key=api_key,
-            base_url="https://openrouter.ai/api/v1"
+            base_url=OPENROUTER_BASE_URL
         )
     return _openrouter_client
 

tools/session_search_tool.py

@@ -23,6 +23,7 @@ import logging
 from typing import Dict, Any, List, Optional
 
 from openai import AsyncOpenAI
+from hermes_constants import OPENROUTER_BASE_URL
 
 SUMMARIZER_MODEL = "google/gemini-3-flash-preview"
 MAX_SESSION_CHARS = 100_000
@@ -40,7 +41,7 @@ def _get_client() -> AsyncOpenAI:
             raise ValueError("OPENROUTER_API_KEY not set")
         _summarizer_client = AsyncOpenAI(
             api_key=api_key,
-            base_url="https://openrouter.ai/api/v1",
+            base_url=OPENROUTER_BASE_URL,
         )
     return _summarizer_client
 

tools/skills_guard.py

@@ -29,6 +29,8 @@ from datetime import datetime, timezone
 from pathlib import Path
 from typing import List, Tuple
 
+from hermes_constants import OPENROUTER_BASE_URL
+
 
 # ---------------------------------------------------------------------------
 # Hardcoded trust configuration
@@ -941,7 +943,7 @@ def llm_audit_skill(skill_path: Path, static_result: ScanResult,
             return static_result
 
         client = OpenAI(
-            base_url="https://openrouter.ai/api/v1",
+            base_url=OPENROUTER_BASE_URL,
             api_key=api_key,
         )
         response = client.chat.completions.create(
@@ -1037,11 +1039,6 @@ def _get_configured_model() -> str:
         return ""
 
 
-def check_guard_requirements() -> Tuple[bool, str]:
-    """Check if the guard module can operate. Always returns True (no external deps)."""
-    return True, "Skills Guard ready"
-
-
 # ---------------------------------------------------------------------------
 # Internal helpers
 # ---------------------------------------------------------------------------

tools/terminal_tool.py

@@ -212,7 +212,7 @@ def _check_disk_usage_warning():
                 if f.is_file():
                     try:
                         total_bytes += f.stat().st_size
-                    except:
+                    except OSError:
                         pass
         
         total_gb = total_bytes / (1024 ** 3)
@@ -341,7 +341,7 @@ def _prompt_dangerous_approval(command: str, description: str, timeout_seconds:
         def get_input():
             try:
                 result["choice"] = input("      Choice [o/s/a/D]: ").strip().lower()
-            except:
+            except (EOFError, OSError):
                 result["choice"] = ""
         
         thread = threading.Thread(target=get_input, daemon=True)
@@ -894,7 +894,7 @@ class _SingularityEnvironment:
         """Cleanup on destruction."""
         try:
             self.cleanup()
-        except:
+        except Exception:
             pass
 
 
@@ -1022,13 +1022,13 @@ class _SSHEnvironment:
                 cmd = ["ssh", "-o", f"ControlPath={self.control_socket}", "-O", "exit", 
                        f"{self.user}@{self.host}"]
                 subprocess.run(cmd, capture_output=True, timeout=5)
-            except:
+            except (OSError, subprocess.SubprocessError):
                 pass
             
             # Remove socket file
             try:
                 self.control_socket.unlink()
-            except:
+            except OSError:
                 pass
     
     def stop(self):
@@ -1039,7 +1039,7 @@ class _SSHEnvironment:
         """Cleanup on destruction."""
         try:
             self.cleanup()
-        except:
+        except Exception:
             pass
 
 
@@ -1112,7 +1112,7 @@ class _DockerEnvironment:
         """Cleanup on destruction."""
         try:
             self.cleanup()
-        except:
+        except Exception:
             pass
 
 
@@ -1189,7 +1189,7 @@ class _ModalEnvironment:
         """Cleanup on destruction."""
         try:
             self.cleanup()
-        except:
+        except Exception:
             pass
 
 
@@ -1504,7 +1504,7 @@ def get_active_environments_info() -> Dict[str, Any]:
                 try:
                     size = sum(f.stat().st_size for f in Path(path).rglob('*') if f.is_file())
                     total_size += size
-                except:
+                except OSError:
                     pass
     
     info["total_disk_usage_mb"] = round(total_size / (1024 * 1024), 2)
@@ -1532,7 +1532,7 @@ def cleanup_all_environments():
         try:
             shutil.rmtree(path, ignore_errors=True)
             print(f"[Terminal Cleanup] Removed orphaned: {path}")
-        except:
+        except OSError:
             pass
     
     if not os.getenv("HERMES_QUIET") and cleaned > 0:
@@ -1669,7 +1669,6 @@ def terminal_tool(
         # This prevents parallel tasks from overwriting each other's files
         # In CLI mode (HERMES_QUIET), use the cwd directly without subdirectories
         if env_type == "local" and not os.getenv("HERMES_QUIET"):
-            import uuid
             with _env_lock:
                 if effective_task_id not in _task_workdirs:
                     task_workdir = Path(cwd) / f"hermes-{effective_task_id}-{uuid.uuid4().hex[:8]}"
@@ -1944,7 +1943,7 @@ def check_terminal_requirements() -> bool:
 
 
 if __name__ == "__main__":
-    """Simple test when run directly."""
+    # Simple test when run directly
     print("Terminal Tool Module (mini-swe-agent backend)")
     print("=" * 50)
     

tools/transcription_tools.py

@@ -99,8 +99,3 @@ def transcribe_audio(file_path: str, model: Optional[str] = None) -> dict:
             "transcript": "",
             "error": str(e),
         }
-
-
-def check_stt_requirements() -> bool:
-    """Check if OpenAI API key is available for speech-to-text."""
-    return bool(os.getenv("HERMES_OPENAI_API_KEY") or os.getenv("OPENAI_API_KEY"))

tools/vision_tools.py

@@ -36,9 +36,9 @@ import base64
 from pathlib import Path
 from typing import Dict, Any, Optional
 from openai import AsyncOpenAI
-import httpx  # Use httpx for async HTTP requests
+import httpx
+from hermes_constants import OPENROUTER_BASE_URL
 
-# Initialize OpenRouter API client lazily (only when needed)
 _openrouter_client = None
 
 def _get_openrouter_client():
@@ -50,7 +50,7 @@ def _get_openrouter_client():
             raise ValueError("OPENROUTER_API_KEY environment variable not set")
         _openrouter_client = AsyncOpenAI(
             api_key=api_key,
-            base_url="https://openrouter.ai/api/v1"
+            base_url=OPENROUTER_BASE_URL
         )
     return _openrouter_client
 

tools/web_tools.py

@@ -50,6 +50,7 @@ from pathlib import Path
 from typing import List, Dict, Any, Optional
 from firecrawl import Firecrawl
 from openai import AsyncOpenAI
+from hermes_constants import OPENROUTER_BASE_URL
 
 # Initialize Firecrawl client lazily (only when needed)
 # This prevents import errors when FIRECRAWL_API_KEY is not set
@@ -77,7 +78,7 @@ def _get_summarizer_client():
             raise ValueError("OPENROUTER_API_KEY environment variable not set")
         _summarizer_client = AsyncOpenAI(
             api_key=api_key,
-            base_url="https://openrouter.ai/api/v1"
+            base_url=OPENROUTER_BASE_URL
         )
     return _summarizer_client
 

toolsets.py

@@ -24,7 +24,6 @@ Usage:
 """
 
 from typing import List, Dict, Any, Set, Optional
-import json
 
 
 # Core toolset definitions
@@ -162,7 +161,7 @@ TOOLSETS = {
     "safe": {
         "description": "Safe toolkit without terminal access",
         "tools": ["mixture_of_agents"],
-        "includes": ["web", "vision", "creative"]
+        "includes": ["web", "vision", "image_gen"]
     },
     
     # ==========================================================================
@@ -587,50 +586,31 @@ def print_toolset_tree(name: str, indent: int = 0) -> None:
 
 
 if __name__ == "__main__":
-    """
+    print("Toolsets System Demo")
-    Demo and testing of the toolsets system
-    """
-    print("🎯 Toolsets System Demo")
     print("=" * 60)
     
-    # Show all available toolsets
+    print("\nAvailable Toolsets:")
-    print("\n📦 Available Toolsets:")
     print("-" * 40)
     for name, toolset in get_all_toolsets().items():
         info = get_toolset_info(name)
-        composite = "📂" if info["is_composite"] else "🔧"
+        composite = "[composite]" if info["is_composite"] else "[leaf]"
-        print(f"{composite} {name:20} - {toolset['description']}")
+        print(f"  {composite} {name:20} - {toolset['description']}")
-        print(f"   Tools: {len(info['resolved_tools'])} total")
+        print(f"     Tools: {len(info['resolved_tools'])} total")
     
-    
+    print("\nToolset Resolution Examples:")
-    # Demo toolset resolution
-    print("\n🔍 Toolset Resolution Examples:")
     print("-" * 40)
-    
+    for name in ["web", "terminal", "safe", "debugging"]:
-    examples = ["research", "development", "full_stack", "minimal", "safe"]
-    for name in examples:
         tools = resolve_toolset(name)
-        print(f"\n{name}:")
+        print(f"\n  {name}:")
-        print(f"  Resolved to {len(tools)} tools: {', '.join(sorted(tools))}")
+        print(f"    Resolved to {len(tools)} tools: {', '.join(sorted(tools))}")
     
-    # Show toolset composition tree
+    print("\nMultiple Toolset Resolution:")
-    print("\n🌳 Toolset Composition Tree:")
     print("-" * 40)
-    print("\nExample: 'content_creation' toolset:")
+    combined = resolve_multiple_toolsets(["web", "vision", "terminal"])
-    print_toolset_tree("content_creation")
+    print(f"  Combining ['web', 'vision', 'terminal']:")
+    print(f"    Result: {', '.join(sorted(combined))}")
     
-    print("\nExample: 'full_stack' toolset:")
+    print("\nCustom Toolset Creation:")
-    print_toolset_tree("full_stack")
-    
-    # Demo multiple toolset resolution
-    print("\n🔗 Multiple Toolset Resolution:")
-    print("-" * 40)
-    combined = resolve_multiple_toolsets(["minimal", "vision", "reasoning"])
-    print(f"Combining ['minimal', 'vision', 'reasoning']:")
-    print(f"  Result: {', '.join(sorted(combined))}")
-    
-    # Demo custom toolset creation
-    print("\n➕ Custom Toolset Creation:")
     print("-" * 40)
     create_custom_toolset(
         name="my_custom",
@@ -638,8 +618,7 @@ if __name__ == "__main__":
         tools=["web_search"],
         includes=["terminal", "vision"]
     )
-    
     custom_info = get_toolset_info("my_custom")
-    print(f"Created 'my_custom' toolset:")
+    print(f"  Created 'my_custom' toolset:")
-    print(f"  Description: {custom_info['description']}")
+    print(f"    Description: {custom_info['description']}")
-    print(f"  Resolved tools: {', '.join(custom_info['resolved_tools'])}")
+    print(f"    Resolved tools: {', '.join(custom_info['resolved_tools'])}")

trajectory_compressor.py

@@ -44,6 +44,7 @@ from datetime import datetime
 import fire
 from rich.progress import Progress, SpinnerColumn, TextColumn, BarColumn, TaskProgressColumn, TimeElapsedColumn, TimeRemainingColumn
 from rich.console import Console
+from hermes_constants import OPENROUTER_BASE_URL
 
 # Load environment variables
 from dotenv import load_dotenv
@@ -70,7 +71,7 @@ class CompressionConfig:
     
     # Summarization (OpenRouter)
     summarization_model: str = "google/gemini-3-flash-preview"
-    base_url: str = "https://openrouter.ai/api/v1"
+    base_url: str = OPENROUTER_BASE_URL
     api_key_env: str = "OPENROUTER_API_KEY"
     temperature: float = 0.3
     max_retries: int = 3