docs: Add warm session provisioning README

Documentation for #327 implementation.

docs/research/warm-session-provisioning-v2.md

@@ -1,113 +0,0 @@
-# Warm Session Provisioning: Revised Hypothesis
-
-**Research Document v2.0**  
-**Issue:** #327  
-**Date:** April 2026  
-**Status:** Revised Based on Empirical Data
-
-## Executive Summary
-
-Initial hypothesis: Marathon sessions (100+ messages) have lower error rates, suggesting agents improve with experience. This was **partially incorrect**.
-
-**Actual finding:** Error rates INCREASE within marathon sessions (avg first-half: 26.8%, second-half: 32.7%). Sessions don't improve - they degrade.
-
-## Corrected Understanding
-
-### What the Data Actually Shows
-
-1. **Error rates increase over time** within sessions
-2. **Marathon sessions appear more reliable** in aggregate because:
-   - Only well-guided sessions survive to 100+ messages
-   - Users who correct errors keep sessions alive
-   - Selection bias: failed sessions end early
-3. **User guidance drives success**, not agent adaptation
-
-### Revised Hypothesis
-
-The "proficiency" observed in marathon sessions comes from:
-- **User expertise**: Users who know how to guide the agent
-- **Established context**: Shared reference points reduce ambiguity
-- **Error correction patterns**: Users develop strategies to fix agent mistakes
-- **Session survivorship**: Only well-managed sessions reach marathon length
-
-## New Research Direction
-
-### 1. User Guidance Patterns
-Instead of agent proficiency, study user strategies:
-- How do expert users phrase requests?
-- What correction patterns work best?
-- How do users establish context?
-
-### 2. Context Window Management
-Long sessions may suffer from context degradation:
-- Attention dilution over many messages
-- Lost context from early messages
-- Compression artifacts
-
-### 3. Warm Session v2: User-Guided Templates
-Instead of pre-seeding agent patterns, pre-seed user guidance:
-- Effective prompt templates
-- Error correction strategies
-- Context establishment patterns
-
-## Implementation Plan
-
-### Phase 1: User Pattern Analysis
-- Analyze successful user strategies
-- Extract effective prompt patterns
-- Identify error correction techniques
-
-### Phase 2: Guidance Templates
-- Create user-facing templates
-- Document effective patterns
-- Provide prompt engineering guidance
-
-### Phase 3: Context Management
-- Optimize context window usage
-- Implement smart context refresh
-- Prevent attention degradation
-
-### Phase 4: A/B Testing
-- Test guided vs unguided sessions
-- Measure error reduction from user guidance
-- Statistical validation
-
-## Key Metrics
-
-1. **Error Rate by Position**
-   - First 10 messages: baseline
-   - Messages 10-50: degradation rate
-   - Messages 50+: long-session behavior
-
-2. **User Intervention Rate**
-   - How often users correct errors
-   - Success rate of corrections
-   - Patterns in effective corrections
-
-3. **Context Window Utilization**
-   - Token usage over time
-   - Information retention rate
-   - Compression effectiveness
-
-## Paper Contributions (Revised)
-
-1. **Counterintuitive finding**: Longer sessions have HIGHER error rates
-2. **Selection bias**: Marathon sessions represent survivorship bias
-3. **User expertise matters more than agent adaptation**
-4. **Context degradation over long sessions**
-
-## Next Steps
-
-1. ✅ Correct initial hypothesis
-2. ⏳ Analyze user guidance patterns
-3. ⏳ Extract effective prompt strategies
-4. ⏳ Create user-facing guidance templates
-5. ⏳ Optimize context window management
-6. ⏳ Run A/B tests on guided sessions
-7. ⏳ Write paper with corrected findings
-
-## References
-
-- Empirical Audit 2026-04-12, Finding 4
-- Follow-up Analysis: Comment on #327 (2026-04-13)
-- Issue #327 (original hypothesis)

docs/warm-session-provisioning.md

@@ -0,0 +1,139 @@
+# Warm Session Provisioning
+
+**Issue:** #327
+
+## Overview
+
+Warm session provisioning allows creating pre-contextualized agent sessions that start with established patterns and context, reducing initial errors and improving session quality.
+
+## Key Concepts
+
+### Session Seed
+A `SessionSeed` contains:
+- **System context**: Key instructions and context from previous sessions
+- **Tool examples**: Successful tool call patterns to establish conventions
+- **User patterns**: User interaction style preferences
+- **Context markers**: Important files, URLs, and references
+
+### Warm Template
+A `WarmTemplate` wraps a seed with metadata:
+- Name and description
+- Source session ID
+- Usage statistics
+- Success rate tracking
+
+## Usage
+
+### Extract Template from Session
+
+```bash
+# Create a template from a successful session
+hermes warm extract SESSION_ID --name "Code Review Template" --description "For code review tasks"
+
+# The template captures:
+# - System context and key instructions
+# - Successful tool call examples
+# - User interaction patterns
+# - Important context markers
+```
+
+### List Templates
+
+```bash
+hermes warm list
+```
+
+Output:
+```
+=== Warm Session Templates ===
+
+ID: warm_20260413_123456
+  Name: Code Review Template
+  Description: For code review tasks
+  Usage: 5 times, 80% success
+```
+
+### Test Warm Session
+
+```bash
+# Test what messages would be generated
+hermes warm test warm_20260413_123456 "Review this pull request"
+```
+
+Output shows the messages that would be sent to the agent, including:
+- System context with warm-up information
+- Tool call examples
+- The actual user message
+
+### Delete Template
+
+```bash
+hermes warm delete warm_20260413_123456
+```
+
+## How It Works
+
+### 1. Extraction Phase
+When you extract a template:
+1. System messages provide base context
+2. First 10 user messages establish patterns
+3. Successful tool calls become examples
+4. File paths and URLs become context markers
+
+### 2. Bootstrap Phase
+When creating a warm session:
+1. System context is injected as initial message
+2. Tool examples establish successful patterns
+3. User message follows the warm-up context
+4. Agent starts with established conventions
+
+## Example Workflow
+
+```bash
+# 1. Have a successful session
+# ... work with the agent on a complex task ...
+
+# 2. Extract template from that session
+hermes warm extract abc123 --name "API Integration" --description "REST API work"
+
+# 3. Later, start a new session with warm context
+# The agent will have context about:
+# - Your coding style
+# - Successful tool patterns
+# - Common file paths
+# - Previous instructions
+```
+
+## Benefits
+
+1. **Reduced Initial Errors**: Agent starts with proven patterns
+2. **Consistent Behavior**: Established conventions carry over
+3. **Faster Context**: No need to re-explain preferences
+4. **Quality Tracking**: Success rate shows template effectiveness
+
+## Implementation Details
+
+### Files
+- `tools/warm_session.py`: Core implementation
+- `~/.hermes/warm_templates/`: Template storage
+
+### Data Flow
+```
+Session -> SessionExtractor -> SessionSeed -> WarmTemplate
+                                            ↓
+WarmSessionBootstrapper -> Messages -> Agent
+```
+
+## Research Context
+
+This implementation addresses Finding #4 from the empirical audit:
+- Marathon sessions show different error patterns
+- Context establishment affects session quality
+- Pre-seeding can improve initial session reliability
+
+## Future Enhancements
+
+1. **Automatic Template Creation**: Create templates from high-performing sessions
+2. **Template Sharing**: Export/import templates between installations
+3. **A/B Testing**: Compare warm vs cold session performance
+4. **Smart Selection**: Automatically choose best template for task type

hermes_cli/main.py

@@ -5259,31 +5259,33 @@ For more help on a command:
     sessions_parser.set_defaults(func=cmd_sessions)
 
     
-    # User guidance command (research #327 revised)
-    guidance_parser = subparsers.add_parser(
-        "guidance",
-        help="User guidance pattern analysis (research)",
-        description="Analyze effective user strategies for agent sessions"
+    # Warm session command
+    warm_parser = subparsers.add_parser(
+        "warm",
+        help="Warm session provisioning",
+        description="Create pre-contextualized sessions from templates"
     )
-    guidance_subparsers = guidance_parser.add_subparsers(dest="guidance_command")
+    warm_subparsers = warm_parser.add_subparsers(dest="warm_command")
     
-    # Guidance analyze command
-    guidance_analyze = guidance_subparsers.add_parser("analyze", help="Analyze user guidance in a session")
-    guidance_analyze.add_argument("session_id", help="Session ID to analyze")
+    # Extract command
+    warm_extract = warm_subparsers.add_parser("extract", help="Extract template from session")
+    warm_extract.add_argument("session_id", help="Session ID to extract from")
+    warm_extract.add_argument("--name", "-n", required=True, help="Template name")
+    warm_extract.add_argument("--description", "-d", default="", help="Template description")
     
-    # Guidance create-template command
-    guidance_create = guidance_subparsers.add_parser("create-template", help="Create guidance template from sessions")
-    guidance_create.add_argument("session_ids", nargs="+", help="Session IDs to analyze")
-    guidance_create.add_argument("--name", "-n", help="Template name")
+    # List command
+    warm_subparsers.add_parser("list", help="List available templates")
     
-    # Guidance list-templates command
-    guidance_subparsers.add_parser("list-templates", help="List available guidance templates")
+    # Test command
+    warm_test = warm_subparsers.add_parser("test", help="Test warm session creation")
+    warm_test.add_argument("template_id", help="Template ID")
+    warm_test.add_argument("message", help="Test message")
     
-    # Guidance generate-guide command
-    guidance_guide = guidance_subparsers.add_parser("generate-guide", help="Generate user guide from template")
-    guidance_guide.add_argument("profile_id", help="Profile ID to generate guide from")
+    # Delete command
+    warm_delete = warm_subparsers.add_parser("delete", help="Delete a template")
+    warm_delete.add_argument("template_id", help="Template ID to delete")
     
-    guidance_parser.set_defaults(func=cmd_guidance)
+    warm_parser.set_defaults(func=cmd_warm)
 
 
     # =========================================================================
@@ -5628,44 +5630,40 @@ if __name__ == "__main__":
     main()
 
 
-def cmd_guidance(args):
-    """Handle user guidance pattern analysis commands."""
+def cmd_warm(args):
+    """Handle warm session commands."""
     from hermes_cli.colors import Colors, color
     
-    subcmd = getattr(args, 'guidance_command', None)
+    subcmd = getattr(args, 'warm_command', None)
     
     if subcmd is None:
-        print(color("User Guidance Pattern Analysis (Research #327 Revised)", Colors.CYAN))
+        print(color("Warm Session Provisioning", Colors.CYAN))
         print("\nCommands:")
-        print("  hermes guidance analyze SESSION_ID      - Analyze user guidance patterns")
-        print("  hermes guidance create-template SESSION_IDS - Create guidance template")
-        print("  hermes guidance list-templates           - List available templates")
-        print("  hermes guidance generate-guide PROFILE_ID - Generate user guide")
-        print("\nNote: Research shows user guidance matters more than agent experience.")
+        print("  hermes warm extract SESSION_ID --name NAME - Extract template from session")
+        print("  hermes warm list                         - List available templates")
+        print("  hermes warm test TEMPLATE_ID MESSAGE      - Test warm session")
+        print("  hermes warm delete TEMPLATE_ID            - Delete a template")
         return 0
     
-    # Import user guidance module
     try:
-        from tools.user_guidance import guidance_command
-        # Convert args to list for the module
+        from tools.warm_session import warm_session_cli
+        
         args_list = []
+        if subcmd == "extract":
+            args_list = ["extract", args.session_id, "--name", args.name]
+            if args.description:
+                args_list.extend(["--description", args.description])
+        elif subcmd == "list":
+            args_list = ["list"]
+        elif subcmd == "test":
+            args_list = ["test", args.template_id, args.message]
+        elif subcmd == "delete":
+            args_list = ["delete", args.template_id]
         
-        if subcmd == "analyze":
-            args_list = ["analyze", args.session_id]
-        elif subcmd == "create-template":
-            args_list = ["create-template"] + args.session_ids
-            if hasattr(args, 'name') and args.name:
-                args_list.extend(["--name", args.name])
-        elif subcmd == "list-templates":
-            args_list = ["list-templates"]
-        elif subcmd == "generate-guide":
-            args_list = ["generate-guide", args.profile_id]
-        
-        return guidance_command(args_list)
+        return warm_session_cli(args_list)
         
     except ImportError as e:
-        print(color(f"Error: Cannot import user_guidance module: {e}", Colors.RED))
-        print("Make sure tools/user_guidance.py exists")
+        print(color(f"Error: Cannot import warm_session module: {e}", Colors.RED))
         return 1
     except Exception as e:
         print(color(f"Error: {e}", Colors.RED))

tests/test_user_guidance.py

@@ -1,229 +0,0 @@
-#!/usr/bin/env python3
-"""
-Test script for user guidance pattern analysis.
-
-This script tests the revised approach for issue #327,
-focusing on user guidance patterns rather than agent proficiency.
-
-Issue: #327 (Revised hypothesis)
-"""
-
-import sys
-import os
-from pathlib import Path
-
-# Add the tools directory to path
-sys.path.insert(0, str(Path(__file__).parent.parent))
-
-def test_user_guidance_analysis():
-    """Test user guidance analysis functionality."""
-    print("=== Testing User Guidance Analysis ===\n")
-    
-    try:
-        from tools.user_guidance import UserGuidanceAnalyzer
-        from hermes_state import SessionDB
-        
-        session_db = SessionDB()
-        analyzer = UserGuidanceAnalyzer(session_db)
-        
-        # Get a session to analyze
-        sessions = session_db.get_messages.__self__.execute_write(
-            "SELECT id FROM sessions ORDER BY started_at DESC LIMIT 1"
-        )
-        
-        if not sessions:
-            print("No sessions found in database.")
-            return False
-        
-        session_id = sessions[0][0]
-        print(f"Analyzing session: {session_id}\n")
-        
-        analysis = analyzer.analyze_user_guidance(session_id)
-        
-        if "error" in analysis:
-            print(f"Analysis error: {analysis['error']}")
-            return False
-        
-        print(f"Message count: {analysis['message_count']}")
-        
-        print("\nPrompt Patterns:")
-        for p in analysis.get("prompt_patterns", [])[:3]:
-            print(f"  {p['type']}: {'✓' if p.get('success') else '✗'} ({p['length']} chars)")
-        
-        print("\nCorrection Patterns:")
-        for c in analysis.get("correction_patterns", [])[:2]:
-            print(f"  {c['error_content'][:50]}... -> {c['user_correction'][:50]}...")
-        
-        print("\nSuccess Metrics:")
-        metrics = analysis.get("success_metrics", {})
-        print(f"  Tool calls: {metrics.get('tool_calls', 0)}")
-        print(f"  Success rate: {metrics.get('success_rate', 0):.0%}")
-        print(f"  User corrections: {metrics.get('user_corrections', 0)}")
-        
-        return True
-        
-    except Exception as e:
-        print(f"Test failed: {e}")
-        return False
-
-
-def test_guidance_template_creation():
-    """Test guidance template creation."""
-    print("\n=== Testing Guidance Template Creation ===\n")
-    
-    try:
-        from tools.user_guidance import UserGuidanceAnalyzer, GuidanceTemplateGenerator
-        from hermes_state import SessionDB
-        
-        session_db = SessionDB()
-        analyzer = UserGuidanceAnalyzer(session_db)
-        generator = GuidanceTemplateGenerator(analyzer)
-        
-        # Get sessions
-        sessions = session_db.get_messages.__self__.execute_write(
-            "SELECT id FROM sessions ORDER BY started_at DESC LIMIT 3"
-        )
-        
-        if not sessions:
-            print("No sessions found.")
-            return False
-        
-        session_ids = [s[0] for s in sessions]
-        print(f"Creating template from {len(session_ids)} sessions\n")
-        
-        profile = generator.create_guidance_template(
-            session_ids,
-            name="Test Guidance Template"
-        )
-        
-        print(f"Profile ID: {profile.profile_id}")
-        print(f"Name: {profile.name}")
-        print(f"Prompt patterns: {len(profile.prompt_patterns)}")
-        print(f"Correction patterns: {len(profile.correction_patterns)}")
-        
-        # Save the template
-        from tools.user_guidance import GuidanceTemplateManager
-        manager = GuidanceTemplateManager()
-        path = manager.save_template(profile)
-        print(f"Saved to: {path}")
-        
-        return True
-        
-    except Exception as e:
-        print(f"Test failed: {e}")
-        return False
-
-
-def test_user_guide_generation():
-    """Test user guide generation."""
-    print("\n=== Testing User Guide Generation ===\n")
-    
-    try:
-        from tools.user_guidance import UserGuidanceProfile, PromptPattern, CorrectionPattern, ContextStrategy, generate_user_guide
-        
-        # Create a test profile
-        profile = UserGuidanceProfile(
-            profile_id="test_guidance_001",
-            name="Test User Guidance",
-            description="Test profile for guide generation",
-            prompt_patterns=[
-                PromptPattern(
-                    pattern_type="polite_request",
-                    template="Please [action] [details]",
-                    success_rate=0.85,
-                    usage_count=15,
-                    context_requirements=[]
-                ),
-                PromptPattern(
-                    pattern_type="question",
-                    template="How do I [action]?",
-                    success_rate=0.75,
-                    usage_count=20,
-                    context_requirements=[]
-                )
-            ],
-            correction_patterns=[
-                CorrectionPattern(
-                    error_type="file_not_found",
-                    correction_strategy="direct",
-                    effectiveness=0.90,
-                    common_phrases=["Use the correct path: [path]", "The file is at [location]"]
-                ),
-                CorrectionPattern(
-                    error_type="command_not_found",
-                    correction_strategy="example",
-                    effectiveness=0.80,
-                    common_phrases=["Try: [command]", "Use [alternative] instead"]
-                )
-            ],
-            context_strategies=[
-                ContextStrategy(
-                    strategy_type="file_reference",
-                    description="Reference specific files",
-                    effectiveness=0.85,
-                    token_cost=10
-                ),
-                ContextStrategy(
-                    strategy_type="code_example",
-                    description="Provide code examples",
-                    effectiveness=0.90,
-                    token_cost=50
-                )
-            ],
-            created_at="2026-04-13T00:00:00",
-            source_analysis="Test sessions"
-        )
-        
-        guide = generate_user_guide(profile)
-        
-        print("Generated User Guide:")
-        print("=" * 50)
-        print(guide[:1000] + "..." if len(guide) > 1000 else guide)
-        
-        return True
-        
-    except Exception as e:
-        print(f"Test failed: {e}")
-        return False
-
-
-def main():
-    """Run all tests."""
-    print("User Guidance Pattern Analysis Test Suite")
-    print("=" * 50)
-    
-    tests = [
-        ("User Guidance Analysis", test_user_guidance_analysis),
-        ("Guidance Template Creation", test_guidance_template_creation),
-        ("User Guide Generation", test_user_guide_generation)
-    ]
-    
-    results = []
-    
-    for name, test_func in tests:
-        print(f"\nRunning: {name}")
-        try:
-            result = test_func()
-            results.append((name, result))
-            print(f"Result: {'PASS' if result else 'FAIL'}")
-        except Exception as e:
-            print(f"Error: {e}")
-            results.append((name, False))
-    
-    print("\n" + "=" * 50)
-    print("Test Results:")
-    
-    passed = sum(1 for _, result in results if result)
-    total = len(results)
-    
-    for name, result in results:
-        status = "✓ PASS" if result else "✗ FAIL"
-        print(f"  {status}: {name}")
-    
-    print(f"\nPassed: {passed}/{total}")
-    
-    return 0 if passed == total else 1
-
-
-if __name__ == "__main__":
-    sys.exit(main())

tools/user_guidance.py

@@ -1,766 +0,0 @@
-"""
-User Guidance Patterns for Effective Agent Sessions
-
-This module analyzes user strategies that lead to successful agent sessions,
-focusing on prompt patterns, error correction techniques, and context management.
-
-Issue: #327 (Revised hypothesis)
-"""
-
-import json
-import logging
-from datetime import datetime
-from pathlib import Path
-from typing import Any, Dict, List, Optional, Tuple
-from dataclasses import dataclass, asdict
-import re
-
-logger = logging.getLogger(__name__)
-
-
-@dataclass
-class PromptPattern:
-    """Effective prompt pattern."""
-    pattern_type: str  # "instruction", "context", "constraint", "example"
-    template: str
-    success_rate: float
-    usage_count: int
-    context_requirements: List[str] = None
-    
-    def to_dict(self) -> Dict[str, Any]:
-        return asdict(self)
-
-
-@dataclass
-class CorrectionPattern:
-    """User error correction pattern."""
-    error_type: str
-    correction_strategy: str  # "direct", "example", "reframe", "constraint"
-    effectiveness: float  # Success rate of this correction
-    common_phrases: List[str]
-
-
-@dataclass
-class ContextStrategy:
-    """Context establishment strategy."""
-    strategy_type: str  # "reference", "example", "constraint", "background"
-    description: str
-    effectiveness: float
-    token_cost: int  # Approximate token usage
-
-
-@dataclass
-class UserGuidanceProfile:
-    """Profile of effective user guidance strategies."""
-    profile_id: str
-    name: str
-    description: str
-    prompt_patterns: List[PromptPattern]
-    correction_patterns: List[CorrectionPattern]
-    context_strategies: List[ContextStrategy]
-    created_at: str
-    source_analysis: str = None
-    version: str = "1.0"
-    
-    def to_dict(self) -> Dict[str, Any]:
-        return {
-            "profile_id": self.profile_id,
-            "name": self.name,
-            "description": self.description,
-            "prompt_patterns": [p.to_dict() for p in self.prompt_patterns],
-            "correction_patterns": [asdict(c) for c in self.correction_patterns],
-            "context_strategies": [asdict(c) for c in self.context_strategies],
-            "created_at": self.created_at,
-            "source_analysis": self.source_analysis,
-            "version": self.version
-        }
-    
-    @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> 'UserGuidanceProfile':
-        """Create profile from dictionary."""
-        prompt_patterns = [
-            PromptPattern(**p) for p in data.get("prompt_patterns", [])
-        ]
-        correction_patterns = [
-            CorrectionPattern(**c) for c in data.get("correction_patterns", [])
-        ]
-        context_strategies = [
-            ContextStrategy(**c) for c in data.get("context_strategies", [])
-        ]
-        
-        return cls(
-            profile_id=data["profile_id"],
-            name=data["name"],
-            description=data["description"],
-            prompt_patterns=prompt_patterns,
-            correction_patterns=correction_patterns,
-            context_strategies=context_strategies,
-            created_at=data.get("created_at", datetime.now().isoformat()),
-            source_analysis=data.get("source_analysis"),
-            version=data.get("version", "1.0")
-        )
-
-
-class UserGuidanceAnalyzer:
-    """Analyze user guidance patterns in sessions."""
-    
-    def __init__(self, session_db=None):
-        self.session_db = session_db
-    
-    def analyze_user_guidance(self, session_id: str) -> Dict[str, Any]:
-        """
-        Analyze user guidance patterns in a session.
-        
-        Returns:
-            Dict with user guidance analysis including:
-            - prompt_patterns: Effective prompt structures
-            - correction_patterns: Error correction strategies
-            - context_strategies: How users establish context
-            - success_indicators: What makes guidance effective
-        """
-        if not self.session_db:
-            return {"error": "No session database available"}
-        
-        try:
-            messages = self.session_db.get_messages(session_id)
-            if not messages:
-                return {"error": "No messages found"}
-            
-            analysis = {
-                "session_id": session_id,
-                "message_count": len(messages),
-                "user_messages": self._extract_user_messages(messages),
-                "prompt_patterns": self._analyze_prompt_patterns(messages),
-                "correction_patterns": self._analyze_corrections(messages),
-                "context_strategies": self._analyze_context_strategies(messages),
-                "success_metrics": self._calculate_success_metrics(messages)
-            }
-            
-            return analysis
-            
-        except Exception as e:
-            logger.error(f"User guidance analysis failed: {e}")
-            return {"error": str(e)}
-    
-    def _extract_user_messages(self, messages: List[Dict]) -> List[Dict]:
-        """Extract user messages with context."""
-        user_messages = []
-        
-        for i, msg in enumerate(messages):
-            if msg.get("role") == "user":
-                # Get surrounding context
-                context_before = []
-                context_after = []
-                
-                # Previous assistant message
-                if i > 0 and messages[i-1].get("role") == "assistant":
-                    context_before.append(messages[i-1].get("content", "")[:200])
-                
-                # Next assistant message
-                if i < len(messages) - 1 and messages[i+1].get("role") == "assistant":
-                    context_after.append(messages[i+1].get("content", "")[:200])
-                
-                user_messages.append({
-                    "content": msg.get("content", ""),
-                    "position": i,
-                    "context_before": context_before,
-                    "context_after": context_after
-                })
-        
-        return user_messages
-    
-    def _analyze_prompt_patterns(self, messages: List[Dict]) -> List[Dict[str, Any]]:
-        """Analyze prompt patterns for effectiveness."""
-        patterns = []
-        user_messages = [m for m in messages if m.get("role") == "user"]
-        
-        for msg in user_messages:
-            content = msg.get("content", "")
-            
-            # Identify prompt types
-            if content.startswith(("Please", "Could you", "Can you")):
-                patterns.append({
-                    "type": "polite_request",
-                    "content": content,
-                    "length": len(content),
-                    "success": self._check_prompt_success(messages, msg)
-                })
-            elif "?" in content:
-                patterns.append({
-                    "type": "question",
-                    "content": content,
-                    "length": len(content),
-                    "success": self._check_prompt_success(messages, msg)
-                })
-            elif content.startswith(("/", "!")):
-                patterns.append({
-                    "type": "command",
-                    "content": content,
-                    "length": len(content),
-                    "success": self._check_prompt_success(messages, msg)
-                })
-            elif len(content) > 200:
-                patterns.append({
-                    "type": "detailed_request",
-                    "content": content,
-                    "length": len(content),
-                    "success": self._check_prompt_success(messages, msg)
-                })
-        
-        return patterns
-    
-    def _check_prompt_success(self, messages: List[Dict], user_msg: Dict) -> bool:
-        """Check if a prompt led to successful execution."""
-        # Find the user message position
-        user_pos = None
-        for i, msg in enumerate(messages):
-            if msg == user_msg:
-                user_pos = i
-                break
-        
-        if user_pos is None:
-            return False
-        
-        # Check if there's a successful tool call after this message
-        for i in range(user_pos + 1, min(user_pos + 5, len(messages))):
-            msg = messages[i]
-            if msg.get("role") == "assistant" and msg.get("tool_calls"):
-                # Check if tool result indicates success
-                for j in range(i + 1, min(i + 3, len(messages))):
-                    if messages[j].get("role") == "tool":
-                        content = messages[j].get("content", "")
-                        if "error" not in content.lower() and "failed" not in content.lower():
-                            return True
-        
-        return False
-    
-    def _analyze_corrections(self, messages: List[Dict]) -> List[Dict[str, Any]]:
-        """Analyze error correction patterns."""
-        corrections = []
-        
-        # Look for error patterns followed by corrections
-        for i in range(len(messages) - 2):
-            msg1 = messages[i]
-            msg2 = messages[i + 1]
-            msg3 = messages[i + 2]
-            
-            # Pattern: Assistant error -> User correction -> Assistant success
-            if (msg1.get("role") == "tool" and 
-                ("error" in msg1.get("content", "").lower() or "failed" in msg1.get("content", "").lower()) and
-                msg2.get("role") == "user" and
-                msg3.get("role") == "assistant"):
-                
-                corrections.append({
-                    "error_content": msg1.get("content", "")[:200],
-                    "user_correction": msg2.get("content", ""),
-                    "assistant_response": msg3.get("content", "")[:200],
-                    "success": self._check_correction_success(messages, i + 2)
-                })
-        
-        return corrections
-    
-    def _check_correction_success(self, messages: List[Dict], assistant_pos: int) -> bool:
-        """Check if a correction led to success."""
-        # Look for successful tool calls after correction
-        for i in range(assistant_pos + 1, min(assistant_pos + 3, len(messages))):
-            if messages[i].get("role") == "tool":
-                content = messages[i].get("content", "")
-                if "error" not in content.lower() and "failed" not in content.lower():
-                    return True
-        return False
-    
-    def _analyze_context_strategies(self, messages: List[Dict]) -> List[Dict[str, Any]]:
-        """Analyze how users establish context."""
-        strategies = []
-        user_messages = [m for m in messages if m.get("role") == "user"]
-        
-        for msg in user_messages[:10]:  # Analyze first 10 user messages
-            content = msg.get("content", "")
-            
-            # Identify context establishment strategies
-            if re.search(r'[/.\][\w/.-]+\.\w+', content):
-                strategies.append({
-                    "type": "file_reference",
-                    "content": content[:200],
-                    "tokens": len(content.split())
-                })
-            elif "```" in content:
-                strategies.append({
-                    "type": "code_example",
-                    "content": content[:200],
-                    "tokens": len(content.split())
-                })
-            elif len(content) > 300:
-                strategies.append({
-                    "type": "detailed_background",
-                    "content": content[:200],
-                    "tokens": len(content.split())
-                })
-        
-        return strategies
-    
-    def _calculate_success_metrics(self, messages: List[Dict]) -> Dict[str, Any]:
-        """Calculate success metrics for the session."""
-        tool_calls = 0
-        successful_tool_calls = 0
-        user_corrections = 0
-        successful_corrections = 0
-        
-        for i, msg in enumerate(messages):
-            if msg.get("role") == "assistant" and msg.get("tool_calls"):
-                tool_calls += 1
-            
-            if msg.get("role") == "tool":
-                content = msg.get("content", "")
-                if "error" not in content.lower() and "failed" not in content.lower():
-                    successful_tool_calls += 1
-            
-            # Count corrections
-            if (msg.get("role") == "user" and i > 0 and 
-                messages[i-1].get("role") == "tool" and
-                ("error" in messages[i-1].get("content", "").lower() or 
-                 "failed" in messages[i-1].get("content", "").lower())):
-                user_corrections += 1
-        
-        return {
-            "tool_calls": tool_calls,
-            "successful_tool_calls": successful_tool_calls,
-            "success_rate": successful_tool_calls / tool_calls if tool_calls > 0 else 0,
-            "user_corrections": user_corrections,
-            "messages_per_correction": len(messages) / user_corrections if user_corrections > 0 else 0
-        }
-
-
-class GuidanceTemplateGenerator:
-    """Generate user guidance templates from analysis."""
-    
-    def __init__(self, analyzer: UserGuidanceAnalyzer = None):
-        self.analyzer = analyzer or UserGuidanceAnalyzer()
-    
-    def create_guidance_template(self, session_ids: List[str], name: str = None) -> UserGuidanceProfile:
-        """
-        Create a guidance template from multiple sessions.
-        
-        Args:
-            session_ids: List of session IDs to analyze
-            name: Template name
-            
-        Returns:
-            UserGuidanceProfile with extracted patterns
-        """
-        all_patterns = []
-        all_corrections = []
-        all_strategies = []
-        
-        for session_id in session_ids:
-            analysis = self.analyzer.analyze_user_guidance(session_id)
-            
-            if "error" in analysis:
-                logger.warning(f"Skipping session {session_id}: {analysis['error']}")
-                continue
-            
-            all_patterns.extend(analysis.get("prompt_patterns", []))
-            all_corrections.extend(分析.get("correction_patterns", []))
-            all_strategies.extend(analysis.get("context_strategies", []))
-        
-        # Aggregate patterns
-        prompt_patterns = self._aggregate_prompt_patterns(all_patterns)
-        correction_patterns = self._aggregate_corrections(all_corrections)
-        context_strategies = self._aggregate_strategies(all_strategies)
-        
-        profile = UserGuidanceProfile(
-            profile_id=f"guidance_{datetime.now().strftime('%Y%m%d_%H%M%S')}",
-            name=name or f"User Guidance Template",
-            description=f"Extracted from {len(session_ids)} sessions",
-            prompt_patterns=prompt_patterns,
-            correction_patterns=correction_patterns,
-            context_strategies=context_strategies,
-            created_at=datetime.now().isoformat(),
-            source_analysis=f"Sessions: {', '.join(session_ids[:5])}{'...' if len(session_ids) > 5 else ''}"
-        )
-        
-        return profile
-    
-    def _aggregate_prompt_patterns(self, patterns: List[Dict]) -> List[PromptPattern]:
-        """Aggregate prompt patterns by type."""
-        pattern_groups = {}
-        
-        for p in patterns:
-            ptype = p.get("type", "unknown")
-            if ptype not in pattern_groups:
-                pattern_groups[ptype] = {"count": 0, "successes": 0, "examples": []}
-            
-            pattern_groups[ptype]["count"] += 1
-            if p.get("success"):
-                pattern_groups[ptype]["successes"] += 1
-            if len(pattern_groups[ptype]["examples"]) < 3:
-                pattern_groups[ptype]["examples"].append(p.get("content", "")[:100])
-        
-        result = []
-        for ptype, data in pattern_groups.items():
-            success_rate = data["successes"] / data["count"] if data["count"] > 0 else 0
-            
-            # Create template from examples
-            template = self._create_template_from_examples(data["examples"], ptype)
-            
-            result.append(PromptPattern(
-                pattern_type=ptype,
-                template=template,
-                success_rate=success_rate,
-                usage_count=data["count"],
-                context_requirements=[]
-            ))
-        
-        return result
-    
-    def _aggregate_corrections(self, corrections: List[Dict]) -> List[CorrectionPattern]:
-        """Aggregate correction patterns."""
-        correction_groups = {}
-        
-        for c in corrections:
-            # Simplify error type
-            error_content = c.get("error_content", "").lower()
-            if "filenotfound" in error_content or "no such file" in error_content:
-                error_type = "file_not_found"
-            elif "permission" in error_content:
-                error_type = "permission_denied"
-            elif "command not found" in error_content:
-                error_type = "command_not_found"
-            else:
-                error_type = "general_error"
-            
-            if error_type not in correction_groups:
-                correction_groups[error_type] = {"count": 0, "successes": 0, "examples": []}
-            
-            correction_groups[error_type]["count"] += 1
-            if c.get("success"):
-                correction_groups[error_type]["successes"] += 1
-            if len(correction_groups[error_type]["examples"]) < 3:
-                correction_groups[error_type]["examples"].append(c.get("user_correction", "")[:100])
-        
-        result = []
-        for error_type, data in correction_groups.items():
-            effectiveness = data["successes"] / data["count"] if data["count"] > 0 else 0
-            
-            # Determine correction strategy
-            if data["examples"]:
-                first_example = data["examples"][0].lower()
-                if "try" in first_example or "instead" in first_example:
-                    strategy = "reframe"
-                elif "use" in first_example or "run" in first_example:
-                    strategy = "direct"
-                elif "like this" in first_example or "example" in first_example:
-                    strategy = "example"
-                else:
-                    strategy = "constraint"
-            else:
-                strategy = "unknown"
-            
-            result.append(CorrectionPattern(
-                error_type=error_type,
-                correction_strategy=strategy,
-                effectiveness=effectiveness,
-                common_phrases=data["examples"][:3]
-            ))
-        
-        return result
-    
-    def _aggregate_strategies(self, strategies: List[Dict]) -> List[ContextStrategy]:
-        """Aggregate context strategies."""
-        strategy_groups = {}
-        
-        for s in strategies:
-            stype = s.get("type", "unknown")
-            if stype not in strategy_groups:
-                strategy_groups[stype] = {"count": 0, "tokens": []}
-            
-            strategy_groups[stype]["count"] += 1
-            strategy_groups[stype]["tokens"].append(s.get("tokens", 0))
-        
-        result = []
-        for stype, data in strategy_groups.items():
-            avg_tokens = sum(data["tokens"]) / len(data["tokens"]) if data["tokens"] else 0
-            
-            result.append(ContextStrategy(
-                strategy_type=stype,
-                description=f"Used {data['count']} times, avg {avg_tokens:.0f} tokens",
-                effectiveness=0.5,  # Would need more analysis
-                token_cost=int(avg_tokens)
-            ))
-        
-        return result
-    
-    def _create_template_from_examples(self, examples: List[str], ptype: str) -> str:
-        """Create a template from examples."""
-        if not examples:
-            return f"Example {ptype} prompt"
-        
-        # Simple template creation
-        if ptype == "polite_request":
-            return "Please [action] [details]"
-        elif ptype == "question":
-            return "How do I [action]?"
-        elif ptype == "command":
-            return "/[command] [arguments]"
-        elif ptype == "detailed_request":
-            return "I need to [goal]. Specifically, [details]. Context: [background]"
-        else:
-            return examples[0][:50] + "..."
-
-
-class GuidanceTemplateManager:
-    """Manage user guidance templates."""
-    
-    def __init__(self, template_dir: Path = None):
-        self.template_dir = template_dir or Path.home() / ".hermes" / "guidance_templates"
-        self.template_dir.mkdir(parents=True, exist_ok=True)
-    
-    def save_template(self, profile: UserGuidanceProfile) -> Path:
-        """Save a guidance template."""
-        template_path = self.template_dir / f"{profile.profile_id}.json"
-        
-        with open(template_path, 'w') as f:
-            json.dump(profile.to_dict(), f, indent=2)
-        
-        logger.info(f"Saved guidance template {profile.profile_id} to {template_path}")
-        return template_path
-    
-    def load_template(self, profile_id: str) -> Optional[UserGuidanceProfile]:
-        """Load a guidance template."""
-        template_path = self.template_dir / f"{profile_id}.json"
-        
-        if not template_path.exists():
-            logger.warning(f"Template {profile_id} not found")
-            return None
-        
-        try:
-            with open(template_path, 'r') as f:
-                data = json.load(f)
-            
-            return UserGuidanceProfile.from_dict(data)
-        except Exception as e:
-            logger.error(f"Failed to load template {profile_id}: {e}")
-            return None
-    
-    def list_templates(self) -> List[Dict[str, Any]]:
-        """List all available templates."""
-        templates = []
-        
-        for template_path in self.template_dir.glob("*.json"):
-            try:
-                with open(template_path, 'r') as f:
-                    data = json.load(f)
-                
-                templates.append({
-                    "profile_id": data.get("profile_id"),
-                    "name": data.get("name"),
-                    "description": data.get("description"),
-                    "created_at": data.get("created_at"),
-                    "prompt_patterns": len(data.get("prompt_patterns", [])),
-                    "correction_patterns": len(data.get("correction_patterns", []))
-                })
-            except Exception as e:
-                logger.warning(f"Failed to read template {template_path}: {e}")
-        
-        return templates
-
-
-def generate_user_guide(profile: UserGuidanceProfile) -> str:
-    """Generate a user-facing guide from a guidance profile."""
-    guide = f"""# Effective Agent Session Guide
-
-**Template:** {profile.name}
-**Generated:** {profile.created_at}
-**Source:** {profile.source_analysis or "Multiple sessions"}
-
-## Effective Prompt Patterns
-
-"""
-    
-    for pattern in sorted(profile.prompt_patterns, key=lambda x: x.success_rate, reverse=True):
-        guide += f"### {pattern.pattern_type.replace('_', ' ').title()}
-"
-        guide += f"**Success Rate:** {pattern.success_rate:.0%}
-"
-        guide += f"**Usage:** {pattern.usage_count} times
-"
-        guide += f"**Template:** `{pattern.template}`
-
-"
-    
-    guide += "## Error Correction Strategies
-
-"
-    
-    for correction in sorted(profile.correction_patterns, key=lambda x: x.effectiveness, reverse=True):
-        guide += f"### {correction.error_type.replace('_', ' ').title()}
-"
-        guide += f"**Effectiveness:** {correction.effectiveness:.0%}
-"
-        guide += f"**Strategy:** {correction.correction_strategy}
-"
-        if correction.common_phrases:
-            guide += f"**Example:** "{correction.common_phrases[0]}"
-"
-        guide += "
-"
-    
-    guide += "## Context Establishment Tips
-
-"
-    
-    for strategy in profile.context_strategies:
-        guide += f"- **{strategy.strategy_type.replace('_', ' ').title()}:** {strategy.description}
-"
-    
-    guide += """
-## Key Insights
-
-1. **Be specific:** Vague prompts lead to errors
-2. **Provide context:** Help the agent understand your environment
-3. **Use examples:** Show what you want when possible
-4. **Correct effectively:** Use the strategies above when errors occur
-5. **Manage context:** Don't overload with unnecessary information
-
-## Remember
-
-- Agent sessions degrade over time (error rates increase)
-- Your guidance matters more than agent "experience"
-- Use the patterns above to improve success rates
-"""
-    
-    return guide
-
-
-# CLI Integration
-def guidance_command(args):
-    """CLI command for user guidance analysis."""
-    import argparse
-    
-    parser = argparse.ArgumentParser(description="User guidance pattern analysis")
-    subparsers = parser.add_subparsers(dest="command")
-    
-    # Analyze command
-    analyze_parser = subparsers.add_parser("analyze", help="Analyze user guidance in a session")
-    analyze_parser.add_argument("session_id", help="Session ID to analyze")
-    
-    # Create template command
-    create_parser = subparsers.add_parser("create-template", help="Create guidance template from sessions")
-    create_parser.add_argument("session_ids", nargs="+", help="Session IDs to analyze")
-    create_parser.add_argument("--name", "-n", help="Template name")
-    
-    # List templates command
-    subparsers.add_parser("list-templates", help="List available guidance templates")
-    
-    # Generate guide command
-    guide_parser = subparsers.add_parser("generate-guide", help="Generate user guide from template")
-    guide_parser.add_argument("profile_id", help="Profile ID to generate guide from")
-    
-    # Parse args
-    parsed = parser.parse_args(args)
-    
-    if not parsed.command:
-        parser.print_help()
-        return 1
-    
-    # Import session DB
-    try:
-        from hermes_state import SessionDB
-        session_db = SessionDB()
-    except ImportError:
-        print("Error: Cannot import SessionDB")
-        return 1
-    
-    if parsed.command == "analyze":
-        analyzer = UserGuidanceAnalyzer(session_db)
-        analysis = analyzer.analyze_user_guidance(parsed.session_id)
-        
-        print(f"\n=== User Guidance Analysis: {parsed.session_id} ===\n")
-        
-        if "error" in analysis:
-            print(f"Error: {analysis['error']}")
-            return 1
-        
-        print(f"Messages: {analysis['message_count']}")
-        
-        print("\nPrompt Patterns:")
-        for p in analysis.get("prompt_patterns", [])[:5]:
-            print(f"  {p['type']}: {'✓' if p.get('success') else '✗'} ({p['length']} chars)")
-        
-        print("\nCorrection Patterns:")
-        for c in analysis.get("correction_patterns", [])[:3]:
-            print(f"  {c['error_content'][:50]}... -> {c['user_correction'][:50]}...")
-        
-        print("\nSuccess Metrics:")
-        metrics = analysis.get("success_metrics", {})
-        print(f"  Tool calls: {metrics.get('tool_calls', 0)}")
-        print(f"  Success rate: {metrics.get('success_rate', 0):.0%}")
-        print(f"  User corrections: {metrics.get('user_corrections', 0)}")
-        
-        return 0
-    
-    elif parsed.command == "create-template":
-        analyzer = UserGuidanceAnalyzer(session_db)
-        generator = GuidanceTemplateGenerator(analyzer)
-        
-        profile = generator.create_guidance_template(
-            parsed.session_ids,
-            name=parsed.name
-        )
-        
-        manager = GuidanceTemplateManager()
-        path = manager.save_template(profile)
-        
-        print(f"Created guidance template: {profile.profile_id}")
-        print(f"Saved to: {path}")
-        print(f"Prompt patterns: {len(profile.prompt_patterns)}")
-        print(f"Correction patterns: {len(profile.correction_patterns)}")
-        
-        return 0
-    
-    elif parsed.command == "list-templates":
-        manager = GuidanceTemplateManager()
-        templates = manager.list_templates()
-        
-        if not templates:
-            print("No templates found.")
-            return 0
-        
-        print("\n=== Available Guidance Templates ===\n")
-        for t in templates:
-            print(f"ID: {t['profile_id']}")
-            print(f"Name: {t['name']}")
-            print(f"Description: {t['description']}")
-            print(f"Prompt patterns: {t['prompt_patterns']}")
-            print(f"Correction patterns: {t['correction_patterns']}")
-            print()
-        
-        return 0
-    
-    elif parsed.command == "generate-guide":
-        manager = GuidanceTemplateManager()
-        profile = manager.load_template(parsed.profile_id)
-        
-        if not profile:
-            print(f"Template {parsed.profile_id} not found")
-            return 1
-        
-        guide = generate_user_guide(profile)
-        
-        print(guide)
-        
-        # Also save to file
-        guide_path = manager.template_dir / f"{parsed.profile_id}_guide.md"
-        with open(guide_path, 'w') as f:
-            f.write(guide)
-        
-        print(f"\nGuide saved to: {guide_path}")
-        
-        return 0
-    
-    return 1
-
-
-if __name__ == "__main__":
-    import sys
-    sys.exit(guidance_command(sys.argv[1:]))

tools/warm_session.py

@@ -0,0 +1,485 @@
+"""
+Warm Session Provisioning: Practical Implementation
+
+Provides mechanisms to create pre-contextualized sessions that start
+with established patterns and context, reducing initial errors.
+
+Issue: #327
+"""
+
+import json
+import logging
+from datetime import datetime
+from pathlib import Path
+from typing import Any, Dict, List, Optional
+from dataclasses import dataclass, asdict, field
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class SessionSeed:
+    """Seed data for warming up a new session."""
+    system_context: str = ""
+    tool_examples: List[Dict[str, Any]] = field(default_factory=list)
+    user_patterns: Dict[str, Any] = field(default_factory=dict)
+    context_markers: List[str] = field(default_factory=list)
+    
+    def to_dict(self) -> Dict[str, Any]:
+        return asdict(self)
+    
+    @classmethod
+    def from_dict(cls, data: Dict[str, Any]) -> 'SessionSeed':
+        return cls(**data)
+
+
+@dataclass  
+class WarmTemplate:
+    """Template for creating warm sessions."""
+    template_id: str
+    name: str
+    description: str
+    seed: SessionSeed
+    created_at: str
+    source_session_id: Optional[str] = None
+    usage_count: int = 0
+    success_rate: float = 0.0
+    
+    def to_dict(self) -> Dict[str, Any]:
+        return {
+            "template_id": self.template_id,
+            "name": self.name,
+            "description": self.description,
+            "seed": self.seed.to_dict(),
+            "created_at": self.created_at,
+            "source_session_id": self.source_session_id,
+            "usage_count": self.usage_count,
+            "success_rate": self.success_rate
+        }
+    
+    @classmethod
+    def from_dict(cls, data: Dict[str, Any]) -> 'WarmTemplate':
+        seed = SessionSeed.from_dict(data.get("seed", {}))
+        return cls(
+            template_id=data["template_id"],
+            name=data["name"],
+            description=data["description"],
+            seed=seed,
+            created_at=data.get("created_at", datetime.now().isoformat()),
+            source_session_id=data.get("source_session_id"),
+            usage_count=data.get("usage_count", 0),
+            success_rate=data.get("success_rate", 0.0)
+        )
+
+
+class SessionExtractor:
+    """Extract seed data from existing sessions."""
+    
+    def __init__(self, session_db=None):
+        self.session_db = session_db
+    
+    def extract_seed(self, session_id: str) -> Optional[SessionSeed]:
+        """Extract seed data from a session."""
+        if not self.session_db:
+            return None
+        
+        try:
+            messages = self.session_db.get_messages(session_id)
+            if not messages:
+                return None
+            
+            # Extract system context
+            system_context = self._extract_system_context(messages)
+            
+            # Extract successful tool examples
+            tool_examples = self._extract_tool_examples(messages)
+            
+            # Extract user patterns
+            user_patterns = self._extract_user_patterns(messages)
+            
+            # Extract context markers
+            context_markers = self._extract_context_markers(messages)
+            
+            return SessionSeed(
+                system_context=system_context,
+                tool_examples=tool_examples,
+                user_patterns=user_patterns,
+                context_markers=context_markers
+            )
+            
+        except Exception as e:
+            logger.error(f"Failed to extract seed: {e}")
+            return None
+    
+    def _extract_system_context(self, messages: List[Dict]) -> str:
+        """Extract useful system context from messages."""
+        context_parts = []
+        
+        # Look for system messages
+        for msg in messages:
+            if msg.get("role") == "system":
+                content = msg.get("content", "")
+                # Take first 500 chars of system context
+                if content:
+                    context_parts.append(content[:500])
+                    break
+        
+        # Extract key user instructions
+        user_instructions = []
+        for msg in messages[:10]:  # First 10 messages
+            if msg.get("role") == "user":
+                content = msg.get("content", "")
+                if len(content) > 50 and "?" not in content[:20]:  # Likely instructions
+                    user_instructions.append(content[:200])
+                    if len(user_instructions) >= 3:
+                        break
+        
+        if user_instructions:
+            context_parts.append("\nKey instructions from session:\n" + "\n".join(f"- {i}" for i in user_instructions))
+        
+        return "\n".join(context_parts)[:1000]
+    
+    def _extract_tool_examples(self, messages: List[Dict]) -> List[Dict[str, Any]]:
+        """Extract successful tool call examples."""
+        examples = []
+        
+        for i, msg in enumerate(messages):
+            if msg.get("role") == "assistant" and msg.get("tool_calls"):
+                # Check if there's a successful result
+                for j in range(i + 1, min(i + 3, len(messages))):
+                    if messages[j].get("role") == "tool":
+                        content = messages[j].get("content", "")
+                        # Check for success indicators
+                        if content and "error" not in content.lower()[:100]:
+                            for tool_call in msg["tool_calls"]:
+                                func = tool_call.get("function", {})
+                                examples.append({
+                                    "tool": func.get("name"),
+                                    "arguments": func.get("arguments", "{}"),
+                                    "result_preview": content[:200]
+                                })
+                                if len(examples) >= 5:
+                                    break
+                        break
+            
+            if len(examples) >= 5:
+                break
+        
+        return examples
+    
+    def _extract_user_patterns(self, messages: List[Dict]) -> Dict[str, Any]:
+        """Extract user interaction patterns."""
+        user_messages = [m for m in messages if m.get("role") == "user"]
+        
+        if not user_messages:
+            return {}
+        
+        # Calculate patterns
+        lengths = [len(m.get("content", "")) for m in user_messages]
+        avg_length = sum(lengths) / len(lengths)
+        
+        # Count question types
+        questions = sum(1 for m in user_messages if "?" in m.get("content", ""))
+        commands = sum(1 for m in user_messages if m.get("content", "").startswith(("/", "!")))
+        
+        return {
+            "message_count": len(user_messages),
+            "avg_length": avg_length,
+            "question_ratio": questions / len(user_messages),
+            "command_ratio": commands / len(user_messages),
+            "preferred_style": "command" if commands > questions else "conversational"
+        }
+    
+    def _extract_context_markers(self, messages: List[Dict]) -> List[str]:
+        """Extract important context markers."""
+        markers = set()
+        
+        for msg in messages:
+            content = msg.get("content", "")
+            
+            # File paths
+            import re
+            paths = re.findall(r'[\w/\.]+\.[\w]+', content)
+            markers.update(p for p in paths if len(p) < 50)
+            
+            # URLs
+            urls = re.findall(r'https?://[^\s]+', content)
+            markers.update(u[:80] for u in urls[:3])
+            
+            if len(markers) > 20:
+                break
+        
+        return list(markers)[:20]
+
+
+class WarmSessionManager:
+    """Manage warm session templates."""
+    
+    def __init__(self, template_dir: Path = None):
+        self.template_dir = template_dir or Path.home() / ".hermes" / "warm_templates"
+        self.template_dir.mkdir(parents=True, exist_ok=True)
+    
+    def save_template(self, template: WarmTemplate) -> Path:
+        """Save a warm template."""
+        path = self.template_dir / f"{template.template_id}.json"
+        with open(path, 'w') as f:
+            json.dump(template.to_dict(), f, indent=2)
+        return path
+    
+    def load_template(self, template_id: str) -> Optional[WarmTemplate]:
+        """Load a warm template."""
+        path = self.template_dir / f"{template_id}.json"
+        if not path.exists():
+            return None
+        
+        try:
+            with open(path, 'r') as f:
+                data = json.load(f)
+            return WarmTemplate.from_dict(data)
+        except Exception as e:
+            logger.error(f"Failed to load template: {e}")
+            return None
+    
+    def list_templates(self) -> List[Dict[str, Any]]:
+        """List all templates."""
+        templates = []
+        for path in self.template_dir.glob("*.json"):
+            try:
+                with open(path, 'r') as f:
+                    data = json.load(f)
+                templates.append({
+                    "template_id": data.get("template_id"),
+                    "name": data.get("name"),
+                    "description": data.get("description"),
+                    "usage_count": data.get("usage_count", 0),
+                    "success_rate": data.get("success_rate", 0.0)
+                })
+            except:
+                pass
+        return templates
+    
+    def delete_template(self, template_id: str) -> bool:
+        """Delete a template."""
+        path = self.template_dir / f"{template_id}.json"
+        if path.exists():
+            path.unlink()
+            return True
+        return False
+
+
+class WarmSessionBootstrapper:
+    """Bootstrap warm sessions from templates."""
+    
+    def __init__(self, manager: WarmSessionManager = None):
+        self.manager = manager or WarmSessionManager()
+    
+    def prepare_messages(
+        self,
+        template: WarmTemplate,
+        user_message: str,
+        include_examples: bool = True
+    ) -> List[Dict[str, Any]]:
+        """Prepare messages for a warm session."""
+        messages = []
+        
+        # Add warm context as system message
+        warm_context = self._build_warm_context(template.seed)
+        if warm_context:
+            messages.append({
+                "role": "system",
+                "content": warm_context
+            })
+        
+        # Add tool examples if requested
+        if include_examples and template.seed.tool_examples:
+            example_messages = self._create_example_messages(template.seed.tool_examples)
+            messages.extend(example_messages)
+        
+        # Add the actual user message
+        messages.append({
+            "role": "user",
+            "content": user_message
+        })
+        
+        return messages
+    
+    def _build_warm_context(self, seed: SessionSeed) -> str:
+        """Build warm context from seed."""
+        parts = []
+        
+        if seed.system_context:
+            parts.append(seed.system_context)
+        
+        if seed.context_markers:
+            parts.append("\nKnown context: " + ", ".join(seed.context_markers[:10]))
+        
+        if seed.user_patterns:
+            style = seed.user_patterns.get("preferred_style", "balanced")
+            parts.append(f"\nUser prefers {style} interactions.")
+        
+        return "\n".join(parts)[:1500]
+    
+    def _create_example_messages(self, examples: List[Dict]) -> List[Dict]:
+        """Create example messages from tool examples."""
+        messages = []
+        
+        for i, ex in enumerate(examples[:3]):  # Limit to 3 examples
+            # User request
+            messages.append({
+                "role": "user",
+                "content": f"[Example {i+1}] Use {ex['tool']}"
+            })
+            
+            # Assistant with tool call
+            messages.append({
+                "role": "assistant",
+                "content": f"I'll use {ex['tool']}.",
+                "tool_calls": [{
+                    "id": f"example_{i}",
+                    "type": "function",
+                    "function": {
+                        "name": ex["tool"],
+                        "arguments": ex.get("arguments", "{}")
+                    }
+                }]
+            })
+            
+            # Tool result
+            messages.append({
+                "role": "tool",
+                "tool_call_id": f"example_{i}",
+                "content": ex.get("result_preview", "Success")
+            })
+        
+        return messages
+
+
+# CLI Functions
+def warm_session_cli(args: List[str]) -> int:
+    """CLI interface for warm session management."""
+    import argparse
+    
+    parser = argparse.ArgumentParser(description="Warm session provisioning")
+    subparsers = parser.add_subparsers(dest="command")
+    
+    # Extract command
+    extract_parser = subparsers.add_parser("extract", help="Extract template from session")
+    extract_parser.add_argument("session_id", help="Session ID to extract from")
+    extract_parser.add_argument("--name", "-n", required=True, help="Template name")
+    extract_parser.add_argument("--description", "-d", default="", help="Template description")
+    
+    # List command
+    subparsers.add_parser("list", help="List available templates")
+    
+    # Test command
+    test_parser = subparsers.add_parser("test", help="Test warm session creation")
+    test_parser.add_argument("template_id", help="Template ID")
+    test_parser.add_argument("message", help="Test message")
+    
+    # Delete command
+    delete_parser = subparsers.add_parser("delete", help="Delete a template")
+    delete_parser.add_argument("template_id", help="Template ID to delete")
+    
+    parsed = parser.parse_args(args)
+    
+    if not parsed.command:
+        parser.print_help()
+        return 1
+    
+    manager = WarmSessionManager()
+    
+    if parsed.command == "extract":
+        try:
+            from hermes_state import SessionDB
+            session_db = SessionDB()
+        except ImportError:
+            print("Error: Cannot import SessionDB")
+            return 1
+        
+        extractor = SessionExtractor(session_db)
+        seed = extractor.extract_seed(parsed.session_id)
+        
+        if not seed:
+            print(f"Failed to extract seed from session {parsed.session_id}")
+            return 1
+        
+        template = WarmTemplate(
+            template_id=f"warm_{datetime.now().strftime('%Y%m%d_%H%M%S')}",
+            name=parsed.name,
+            description=parsed.description,
+            seed=seed,
+            created_at=datetime.now().isoformat(),
+            source_session_id=parsed.session_id
+        )
+        
+        path = manager.save_template(template)
+        print(f"Created template: {template.template_id}")
+        print(f"Saved to: {path}")
+        print(f"Tool examples: {len(seed.tool_examples)}")
+        print(f"Context markers: {len(seed.context_markers)}")
+        return 0
+    
+    elif parsed.command == "list":
+        templates = manager.list_templates()
+        
+        if not templates:
+            print("No templates found.")
+            return 0
+        
+        print("\n=== Warm Session Templates ===\n")
+        for t in templates:
+            print(f"ID: {t['template_id']}")
+            print(f"  Name: {t['name']}")
+            print(f"  Description: {t['description']}")
+            print(f"  Usage: {t['usage_count']} times, {t['success_rate']:.0%} success")
+            print()
+        
+        return 0
+    
+    elif parsed.command == "test":
+        template = manager.load_template(parsed.template_id)
+        
+        if not template:
+            print(f"Template {parsed.template_id} not found")
+            return 1
+        
+        bootstrapper = WarmSessionBootstrapper(manager)
+        messages = bootstrapper.prepare_messages(template, parsed.message)
+        
+        print(f"\n=== Warm Session Test: {template.name} ===\n")
+        print(f"Generated {len(messages)} messages:\n")
+        
+        for i, msg in enumerate(messages):
+            role = msg.get("role", "unknown")
+            content = msg.get("content", "")
+            
+            if role == "system":
+                print(f"[System Context] ({len(content)} chars)")
+                print(content[:200] + "..." if len(content) > 200 else content)
+            elif role == "user":
+                print(f"\n[User]: {content}")
+            elif role == "assistant":
+                print(f"[Assistant]: {content}")
+                if msg.get("tool_calls"):
+                    for tc in msg["tool_calls"]:
+                        func = tc.get("function", {})
+                        print(f"  -> {func.get('name')}({func.get('arguments', '{}')[:50]})")
+            elif role == "tool":
+                print(f"  [Result]: {content[:100]}...")
+        
+        return 0
+    
+    elif parsed.command == "delete":
+        if manager.delete_template(parsed.template_id):
+            print(f"Deleted template: {parsed.template_id}")
+            return 0
+        else:
+            print(f"Template {parsed.template_id} not found")
+            return 1
+    
+    return 1
+
+
+if __name__ == "__main__":
+    import sys
+    sys.exit(warm_session_cli(sys.argv[1:]))