Is Sibyl Planner user-invocable?

Based on its configuration, it is not directly user-invocable and is intended to be called as a sub-skill by the Sibyl orchestration system.

How does the planning mode work?

It uses pilot configuration summaries and Sibyl's orchestration logic to generate a detailed, structured plan for research or development tasks.

What is the primary purpose of Sibyl Planner?

Sibyl Planner is designed to create rigorous, reproducible experimental plans and troubleshoot GPU issues within the Sibyl framework.

Does this skill support GPU debugging?

Yes, it includes a specific 'fix-gpu' mode to identify and resolve issues with GPU infrastructure and drivers.

What tools does the Sibyl Planner utilize?

It has access to Read, Write, Bash, WebSearch, and WebFetch to gather data, modify files, and implement experimental plans.

Sibyl Experiment Planner

Name: Sibyl Experiment Planner
Author: qcai0427

byqcai0427

0•

Data Science & ML

Generates rigorous, reproducible experimental plans and resolves GPU infrastructure issues for data science workflows.

Sibyl Experiment Planner is a specialized agent designed to streamline the research and development lifecycle by orchestrating rigorous experimental protocols. It assists users in designing structured experiment plans that ensure reproducibility, while also providing automated troubleshooting for common GPU-related failures. By leveraging the Sibyl system's orchestration capabilities, it helps developers and researchers move from conceptual pilot configurations to full-scale, well-documented experimental execution with consistent results.

Key Features

01Orchestrates complex workflows using the Sibyl-standard agent tier

02Leverages web search and fetching for up-to-date research best practices

03Automates troubleshooting for GPU-related hardware and software issues

04Integrates pilot configuration summaries into comprehensive plans

050 GitHub stars

06Designs rigorous and reproducible experimental protocols

Use Cases

01Debugging and fixing CUDA or GPU driver conflicts in experimental environments

02Standardizing experimental documentation for research teams to ensure consistency

03Designing a systematic benchmarking plan for new machine learning models

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add qcai0427/sibyl-codex sibyl-planner

For use in Claude.ai and ChatGPT