Home

Welcome to the Aludel wiki!

Aludel is a Phoenix-based workbench for refining prompts, comparing model behavior, and tracing how each revision alters the final alloy of quality, latency, and cost.

flowchart LR
    A[Design Prompt] --> B[Run Across Providers]
    B --> C[Review Responses]
    C --> D[Validate With Test Suites]
    D --> E[Track Cost, Latency, and Version History]

Loading

What Is Forged Here

• Multi-provider prompt execution from one workspace
• Versioned prompt templates with {{variable}} interpolation
• Repeatable evaluation suites with assertion-based validation
• Separate typed project groupings for prompts and suites
• PostgreSQL-backed metrics for latency, tokens, and estimated cost

The Workbench

Get Started

• Installation - How to light the furnace: standalone, embedded, and Docker deployment paths
• Quick Start - The first working ritual, from provider setup to results
• Providers - The available reagents: OpenAI, Anthropic, and Ollama

Understand the System

• Architecture - The chambers, conduits, and execution flow beneath the surface
• Data Model - Core entities, relationships, and storage choices
• Prompts - Template syntax, versioning, and the prompt lifecycle
• Roadmap - Milestones, priorities, and the next stages of the workbench

Suggested Path

1. Installation
2. Quick Start
3. Providers
4. Architecture
5. Data Model

Community

• Issues
• Discussions