This directory contains planning documents, research notes, and decision logs to maintain context across development sessions.

The .context/ directory serves as a persistent knowledge base for Reflex development, enabling:

• Continuity across development sessions (human and AI)
• Decision tracking and rationale
• Research caching to avoid duplicate work
• Onboarding for new contributors
• AI assistant collaboration and handoff

Primary task tracking and implementation roadmap

Contains:

• Executive summary of project status
• MVP goals and success criteria
• Task breakdown by module (P0/P1/P2/P3 priorities)
• Implementation phases and timeline
• Open questions and design decisions
• Performance targets
• Maintenance strategy

Usage:

• Read at start of every session
• Update task statuses as work progresses
• Add new tasks as discovered
• Document architectural decisions

Create *_RESEARCH.md files as needed to cache important findings:

• TREE_SITTER_RESEARCH.md - Grammar investigation, node types, query patterns
• PERFORMANCE_RESEARCH.md - Benchmarks, optimizations, bottleneck analysis
• BINARY_FORMAT_RESEARCH.md - Serialization design, format comparisons
• LANGUAGE_SPECIFIC_NOTES.md - Per-language implementation details
• {TOPIC}_RESEARCH.md - Any other focused investigation

See the "Context Management & AI Workflow" section in CLAUDE.md for detailed instructions.

1. Start session: Read CLAUDE.md → Read TODO.md
2. During work: Update TODO.md task statuses, create/update RESEARCH.md files
3. End session: Ensure all statuses accurate, document blockers
4. Research: Create focused RESEARCH.md files with examples and version numbers

• TODO.md - Main task tracker (required)
• {TOPIC}_RESEARCH.md - Research findings (uppercase, descriptive)
• README.md - This file (explains the directory)

All .context/ files should be committed to git. They are part of the project documentation and help maintain development continuity.

# Tree-sitter Rust Grammar Research

**Last Updated:** 2025-10-31
**Grammar Version:** tree-sitter-rust 0.23

## Node Types for Symbol Extraction

### Functions
- Node kind: `function_item`
- Name field: `name` (identifier)
- Parameters: `parameters` (parameter_list)
- Example AST: ...

### Structs
- Node kind: `struct_item`
- Fields: ...

## Edge Cases

### Procedural Macros
...

## References
- https://github.com/tree-sitter/tree-sitter-rust
- Grammar docs: ...

#### P0: Tree-sitter Integration (CRITICAL PATH)
- [x] Set up Tree-sitter grammar dependencies (Line 83)
- [x] Add tree-sitter-rust to Cargo.toml - COMPLETED 2025-10-31
- [ ] Add tree-sitter-python to Cargo.toml
- [in_progress] Implement Rust parser (src/parsers/rust.rs)
  - Started: 2025-10-31
  - Status: Completed basic node traversal, working on macro handling
  - See: .context/RUST_PARSER_RESEARCH.md for findings
  - Blocker: Need to understand proc_macro AST representation

• Keep TODO.md up to date with actual project state
• Archive completed RESEARCH.md files or integrate findings into docs
• Update timestamps when making significant changes
• Cross-reference between TODO.md and RESEARCH.md files

Remember: The .context/ directory is your memory between sessions. Use it well!