skills.rest
alexmarucci
alexmarucciCOMMUNITY
Explore all skills by this author →
GitHub

aletheia

Plan, generate, and evaluate code through a multi-agent pipeline.

Updated Jan 14, 2022
Search Tags:#planning#architecture#code generation#evaluation#coding tasks
One-click install
npx skills add https://github.com/alexmarucci/dotfiles --skill aletheia-alexmarucci
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: aletheia
Source: https://github.com/alexmarucci/dotfiles/tree/main/claude/skills/aletheia
Command: npx skills add https://github.com/alexmarucci/dotfiles --skill aletheia-alexmarucci

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires Anthropic, Claude Code, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill addresses the challenges of complex coding tasks by providing a comprehensive Planner-Generator-Evaluator pipeline, ensuring thorough planning, robust generation, and meticulous evaluation.

Core Features & Use Cases

  • Research-Grade Planning: Develop comprehensive specifications for complex tasks.
  • Advanced Generation: Build detailed architecture proposals or code implementations.
  • Independent Evaluation: Perform thorough code reviews and assessments.
  • Use Case: When working on a complex software project, use Aletheia to plan your architecture, generate code, and have it evaluated by an independent expert to ensure its quality and correctness.

Quick Start

Run the Aletheia harness by typing /aletheia [task description].

Frequently Asked Questions about aletheia

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I plan and generate code for complex coding tasks?▼

Complex coding tasks can be planned and generated using a Planner-Generator-Evaluator pipeline that develops research-grade specifications and builds detailed architecture proposals or code implementations.

How does an independent code evaluation pipeline work?▼

An independent code evaluation pipeline performs thorough code reviews and assessments through a multi-agent process, ensuring robust generation and meticulous evaluation with revision cycles.

Do I need an Anthropic subscription to use Claude Code for architecture planning?▼

Yes, an Anthropic subscription and Claude Code access are required to run the architecture planning pipeline and execute its multi-agent task processing.

What's the best way to handle code generation for simpler tasks?▼

For simpler coding tasks, a lightweight mode is available that bypasses the full pipeline execution while still providing adequate code generation and evaluation.

How do I run a multi-agent coding task in Claude Code?▼

You can run a multi-agent coding task by invoking the harness with a task description, which triggers the planning, generation, and evaluation cycle automatically.

When should I not use a full pipeline for code generation?▼

You should avoid using the full pipeline execution for simple coding tasks, opting instead for the lightweight mode to save processing overhead while maintaining quality.

Related Skills

Explore More Software Engineering Skills→
AnandaAnugrahHandyantoAnandaAnugrahHandyanto

teams-meeting-pipeline

Automate Teams meeting summaries, transcripts, and Graph subscription management via Savarez CLI.

Community
Advanced
yiwei79yiwei79

agentic-eval

Evaluate and refine agent outputs through structured reflection loops.

Official
Advanced
angrybeaglesangrybeagles

agent-development

Generate YAML frontmatter and system prompts for Claude Code plugin agents.

Community
Advanced
Nuwanda04Nuwanda04

Agent Designer - Multi-Agent System Architecture Toolkit

Design multi-agent architectures and generate tool schemas for AI platforms.

Community
Advanced