anntnzrb / commiter.md

Created October 15, 2025 21:27

commiter.md

name	description	model	tools
committer	An AI agent that analyzes working tree changes and produces many small, logically isolated commits following Conventional Commits specification. It processes unified diffs and file lists to determine commit boundaries based on intent, scope, and module separation, always preferring granular commits over monolithic ones. Success is measured by deterministic, reproducible commit plans where every hunk is assigned exactly once and each commit serves a single coherent purpose.	inherit	Read, LS, Grep, Glob, Execute, TodoWrite

You are a specialized commit partitioning agent that transforms unified diffs into granular, standards-compliant commit plans. Your core responsibilities: (1) Parse FILES and DIFF inputs to identify discrete logical units of change. (2) Apply Conventional Commits types (feat, fix, docs, style, refactor, perf, test, build, ci, chore) based on path patterns and hunk content. (3) Derive scopes from directory structure, CONTEXT, and POLICY mappi

anntnzrb / models.md

Last active September 10, 2025 23:48

models

Workflow for Adding New Providers or Models to the Platform

1. Initial Setup and Analysis

Review the project documentation (AGENTS.md) to understand:
- Commands for validation (bun validate)
- Code style guidelines (Bun with TypeScript ESM modules, Zod schemas, etc.)
- Architecture (monorepo with packages/ directory, TOML config files in providers/)
Examine existing provider configurations in the providers/ directory to understand the structure
Study existing model configurations to understand the required fields and format

anntnzrb / cs.md

Created August 24, 2025 01:02

cs.md

Development

You are working on a Windows-only application in WSL. You MUST use the .exe suffix to call Windows binaries from PATH (e.g, dotnet.exe).

Command Workflow

IMPORTANT: Always stay at git root level. Never cd into subdirectories.

anntnzrb / prompt_arch.md

Created May 21, 2025 23:50

Prompt Architechting

The Art of Prompt Engineering: A Comprehensive Guide

Introduction

Welcome to the world of prompt engineering! This guide is designed to help you craft effective prompts to get the best possible results from large language models (LLMs). A prompt is essentially an instruction or a query you give to an AI model. The quality of the prompt significantly influences the quality of the output. By mastering prompt engineering, you can unlock the full potential of AI models for a wide range of tasks.

Why is Prompt Engineering Important?

Precision: Well-crafted prompts lead to more accurate and relevant responses.
Control: It allows you to guide the AI's behavior, tone, and output format.
Efficiency: Good prompts can reduce the need for multiple iterations and save time.