prompt-contract

# Prompt Contracts

> A formal framework for managing LLM context and logic through structured "Contracts" and "Cognitive Archetypes."

This project provides a professional alternative to "Vibe Coding" by treating LLM interactions as governed system architecture. It includes foundational theory, structured contract layers, and execution protocols designed for deterministic AI engineering.

## Core Framework
- [Foundation](/1_Foundation): Theory of Intelligence Containment and why models hallucinate.
- [Contract Structure](/2_Contract_Structure): Detailed specifications for Scope, Role, Objective, Reasoning, and Language layers.
- [Execution Protocols](/3_Execution_Protocols): High-reliability workflows including Spec-Driven Development, Task Decomposition, and TDD.

## Implementation & Tools
- [Model Mapping](/4_Model_Mapping): Framework for assigning tasks to Reasoning-Dense vs. Instruction-Following models.
- [Master Templates](/5_Examples): Production-ready contract templates for daily development (Feature Dev, Debugging, PR Review).

## Best Practices
- **Order Matters:** Place Role and Language contracts at the top to optimize for **Prompt Caching**.
- **Version Locking:** Always pin specific model snapshots to prevent instruction drift.
- **Context Hygiene:** Use the `Context_Reset_Protocol` to purge "context rot" and maintain high reasoning quality.

**Maintainer:** Suhail T (https://github.com/m3dcodie)
**License:** MIT