[FEATUE REQUEST] 200K is not enough. Please bring 1M token.

#
## Environment

• [x] Anthropic API
• [ ] AWS Bedrock
• [ ] Google Vertex AI
• [ ] Other: <!-- specify -->
• Claude CLI version: 1.0.93
• Operating System: macOS
• Terminal: Claude Code

## Feature Request
Context window hard limit of 200K tokens severely restricts complex project workflows requiring extensive documentation, MCP tool
definitions, and multi-session state management.

## Current Limitation

• Hard limit: 200K tokens (110% = critical failure)
• MCP tools alone: 159.6K tokens (79.8% of total)
• Memory files: 19.5K tokens (9.8%)
• Remaining for actual work: ~15K tokens

## Use Case
Professional n8n workflow engineering with:

• Extensive business requirements documentation
• Complex MCP tool integrations (n8n, Supabase, GitHub, etc.)
• Multi-session project memory persistence
• Large codebase analysis and validation

## Requested Enhancement
Increase context window options to:

• 1M tokens (primary request)
• 500K tokens (minimum viable)
• User-configurable limits per project

## Business Impact
Current 200K limit forces constant /compact operations, breaking:

• Deep technical analysis workflows
• Multi-step validation processes
• Project continuity across sessions
• Professional workflow engineering tasks

## Similar Tools

• Gemini 1.5 Pro: 1M+ tokens ✅

## Expected Behavior
Context usage should scale to project complexity without forced interruptions at 200K token ceiling.

## Current Workaround
Frequent /compact operations with external memory storage - inefficient for professional workflows requiring continuous context.