Skip to content
Prompt

WitsV3 — Cursor Rules

by capnknives

AI Summary

Cursor Rules for WitsV3 provides development guidelines and best practices for contributing to an LLM orchestration system, covering architecture, async patterns, code structure, testing, and task management. Developers working on WitsV3 or similar LLM wrapper projects benefit from these structured conventions.

Install

Copy this and paste it into Claude Code, Cursor, or any AI assistant:

I want to add the "WitsV3 — Cursor Rules" prompt rules to my project.
Repository: https://github.com/capnknives/WitsV3

Please read the repo to find the rules/prompt file, then:
1. Download it to the correct location (.cursorrules, .windsurfrules, .github/prompts/, or project root — based on the file type)
2. If there's an existing rules file, merge the new rules in rather than overwriting
3. Confirm what was added

Description

Cursor Rules for WitsV3

🧪 Testing Requirements

• Create pytest tests for all new features in /tests mirroring structure • Test async functions with pytest-asyncio • Mock external services (Ollama, file system) in tests • Minimum coverage per feature: • Happy path test • Error handling test • Edge case test • Run existing test functions at bottom of modules when updating

🔄 Project Context & Architecture

• Always read PLANNING.md at the start of new conversations to understand WitsV3's LLM wrapper architecture and goals • Check TASK.md before starting work. Add tasks if not listed with date • WitsV3 is an LLM orchestration system with CLI-first approach, ReAct pattern, and tool registry • Core components: agents/, core/, tools/ - maintain this structure

🐍 Python & Async Best Practices

• Use Python 3.10+ with type hints throughout • All agent/tool methods must be async - WitsV3 is fully async • Follow PEP8 and format with black • Use Pydantic models for data validation (already in core/schemas.py) • Import style: Use relative imports within packages, absolute for cross-package

🧱 Code Structure & Modularity

• Max 500 lines per file - split into modules when approaching limit • Maintain existing structure: • agents/ - Agent implementations (BaseAgent, Orchestrators, Control Center) • core/ - Core functionality (config, LLM interface, memory, schemas) • tools/ - Tool implementations extending BaseTool • New tools go in tools/ and must extend BaseTool • New agents go in agents/ and must extend BaseAgent

Discussion

0/2000
Loading comments...

Health Signals

MaintenanceCommitted 10mo ago
Stale
AdoptionUnder 100 stars
0 ★ · Niche
DocsMissing or thin
Undocumented

GitHub Signals

Issues1
Updated10mo ago
View on GitHub
MIT License

My Fox Den

Community Rating

Sign in to rate this booster

Works With

Any AI assistant that accepts custom rules or system prompts

Claude
ChatGPT
Cursor
Windsurf
Copilot
+ more