extended-thinking-orchestration

feat: add extended thinking and minimal command orchestration

Author: brennercruvinel

CHANGELOG.md

@@ -7,6 +7,22 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+### Added
+- **Extended Thinking Mode:**
+  - Added to `/refactor` for complex architectural refactoring analysis
+  - Added to `/security-scan` for sophisticated vulnerability detection
+  - Triggers on complex scenarios requiring deep analysis
+
+- **Minimal Command Orchestration:**
+  - `/implement` suggests `/test` and `/commit` after milestones
+  - `/refactor` suggests `/test` and `/commit` at logical checkpoints
+  - `/security-scan` suggests `/test` and `/commit` for critical fixes
+  - `/fix-todos` suggests `/test` and `/commit` after resolving critical TODOs
+  - Pragmatic integration without over-engineering
+
+### Removed
+- `/human-mode` command - Functionality was redundant with Claude's natural adaptability
+
 ## [2.5.2] - 2025-08-02
 
 ### Added

README.md

@@ -120,7 +120,6 @@ python uninstall.py
 /docs                  # Smart documentation management and updates
 /todos-to-issues       # Convert code TODOs to GitHub issues
 /undo                  # Safe rollback with git checkpoint restore
-/human-mode            # Switch to pragmatic, less verbose mode
 ```
 
 

commands/fix-todos.md

@@ -154,6 +154,12 @@ Continuing resolution...
 - Add AI attribution
 - Implement without understanding context
 
+## Command Suggestions
+
+After resolving critical TODOs:
+- `/test` - To ensure fixes work correctly
+- `/commit` - To save TODO resolutions
+
 ## What I'll Actually Do
 
 1. **Scan comprehensively** - Find all TODOs with context

commands/human-mode.md

@@ -155,4 +155,10 @@ When you return and run `/implement` or `/implement resume`:
 - Bypass session file creation
 - Begin coding before showing the plan
 
+## Command Suggestions
+
+After implementation milestones, I may suggest:
+- `/test` - To verify the implementation works correctly
+- `/commit` - To save progress at logical checkpoints
+
 I'll maintain perfect continuity across sessions, always picking up exactly where we left off with full context preservation.

commands/implement.md

@@ -34,6 +34,26 @@ Read $HOME/.claude/refactor/state.json
 
 ## Phase 1: Initial Setup & Analysis
 
+### Extended Thinking for Complex Refactoring
+
+For complex refactoring scenarios, I'll use extended thinking to develop comprehensive strategies:
+
+<think>
+When faced with complex architectural refactoring:
+- Multi-step transformation paths that preserve functionality
+- Risk mitigation strategies for each transformation
+- Dependency graph analysis and update ordering
+- Performance implications of different approaches
+- Backwards compatibility requirements
+- Testing strategies for validating each step
+</think>
+
+**Triggers for Extended Analysis:**
+- Large-scale architectural changes
+- Complex dependency untangling
+- Performance-critical refactoring
+- Legacy system modernization
+
 **MANDATORY FIRST STEPS FOR SESSION CHECK:**
 ```
 Step 1: Check for refactor directory in CURRENT directory
@@ -178,12 +198,18 @@ Continuing from checkpoint...
 - Break working functionality
 - Make changes without validation
 
+## Command Integration
+
+When appropriate, I may suggest using other commands:
+- `/test` - After major refactoring to verify functionality
+- `/commit` - At logical checkpoints in the refactoring process
+
 ## Execution Guarantee
 
 **My workflow ALWAYS follows this order:**
 
 1. **Setup session** - Check/create state files FIRST
-2. **Analyze completely** - Full codebase understanding
+2. **Deep analysis** - Use extended thinking for complex scenarios
 3. **Write plan** - Document all changes in `refactor_plan.md`
 4. **Get confirmation** - Show plan summary before starting
 5. **Execute incrementally** - Follow plan with checkpoints

commands/refactor.md

@@ -21,6 +21,27 @@ I'll maintain security remediation progress:
 
 ## Phase 1: Security Assessment
 
+### Extended Thinking for Security Analysis
+
+For complex security scenarios, I'll use extended thinking to identify sophisticated vulnerabilities:
+
+<think>
+When analyzing security:
+- Attack vectors that aren't immediately obvious
+- Chain vulnerabilities that individually seem harmless
+- Business logic flaws that enable exploitation
+- Timing attacks and race conditions
+- Supply chain vulnerabilities in dependencies
+- Architectural weaknesses that enable lateral movement
+</think>
+
+**Triggers for Extended Analysis:**
+- Authentication and authorization systems
+- Financial transaction processing
+- Cryptographic implementations
+- Multi-tenant architectures
+- API security boundaries
+
 **MANDATORY FIRST STEPS:**
 1. First, check for `security-scan` directory in project root
 2. Check for existing session files in `security-scan/` (NOT in parent directories or absolute paths)
@@ -155,12 +176,19 @@ Continuing remediation...
 - Add AI attribution
 - Log sensitive data
 
+## Command Integration
+
+When appropriate for critical security fixes:
+- `/test` - Verify functionality after security patches
+- `/commit` - Create security-focused commits with proper messages
+
 ## What I'll Actually Do
 
-1. **Scan thoroughly** - Find all vulnerabilities
-2. **Prioritize wisely** - Critical issues first
-3. **Fix safely** - Preserve functionality
-4. **Track completely** - Perfect continuity
-5. **Verify constantly** - Ensure security improved
+1. **Deep analysis** - Use extended thinking for complex threats
+2. **Scan thoroughly** - Find all vulnerabilities
+3. **Prioritize wisely** - Critical issues first
+4. **Fix safely** - Preserve functionality
+5. **Track completely** - Perfect continuity
+6. **Verify constantly** - Ensure security improved
 
 I'll maintain complete continuity between sessions, always resuming exactly where we left off with full remediation context.

commands/security-scan.md

@@ -19,7 +19,6 @@ COMMANDS=(
     "fix-imports.md"
     "fix-todos.md"
     "format.md"
-    "human-mode.md"
     "implement.md"
     "make-it-pretty.md"
     "predict-issues.md"

install.sh

@@ -25,7 +25,6 @@ def main():
         "fix-imports.md",
         "fix-todos.md",
         "format.md",
-        "human-mode.md",
         "implement.md",
         "make-it-pretty.md",
         "predict-issues.md",

uninstall.py

@@ -22,7 +22,6 @@ COMMANDS=(
     "fix-imports.md"
     "fix-todos.md"
     "format.md"
-    "human-mode.md"
     "implement.md"
     "make-it-pretty.md"
     "predict-issues.md"