Skill UI

A mechanism for spawning hierarchical, nested sub-agents up to five levels deep, designed specifically for deep task delegation while managing context limitations. Instead of parallel execution, this ensures that each layer of decomposition receives a fresh context window, preventing the parent agent's context budget from being overwhelmed by complex, multi-step workflows. Ideal for research, verification, and synthesis processes.

This tool is designed to help optimize and refine prompts used with Large Language Models (LLMs). By structuring inputs, it ensures that the instructions given to AI systems are clear, comprehensive, and contextually rich, resulting in significantly higher quality and relevance of the generated output. Useful for developers, writers, and researchers.

A utility designed to generate high-dimensional vector embeddings (384-dim) from various sources including text, code, or documents. This process converts raw data into numerical representations optimized for semantic search, similarity comparison, and clustering, utilizing ONNX and HNSW indexing for efficient retrieval.

A powerful skill utilizing the Google Gemini API to generate high-quality images from text prompts or edit existing photos. It supports various aspect ratios, resolutions (up to 4K), and offers model selection for optimal results, making it ideal for creative design, conceptual visualization, and professional visual content creation.

Share durable context and handoffs between AI agents like Claude and Codex via the local ECC Memory Vault. Use this skill to save work state, transfer context, resume tasks, or search shared project knowledge across different agent harnesses.

A specialized guide for authors targeting the ACM Conference on Fairness, Accountability, and Transparency (FAccT). It helps authors strategically frame their responsible AI research, ensuring claims are rigorously grounded in both technical measurement and social context. Covers necessary evidence bars, scope calibration, and submission best practices for this niche venue.

This comprehensive guide assists researchers in converting an accepted, anonymous draft for the Conference on Learning Theory (COLT) into the final, polished camera-ready version for publication in PMLR proceedings. It covers critical steps such as de-anonymization (restoring author details, acknowledgements), consolidating all promised revisions into the main text, verifying PMLR metadata and BibTeX consistency, and preparing the full archival PDF.

A comprehensive pre-upload audit checklist for submitting papers to the CoRL (Conference on Robot Learning). This guide details specific requirements for the submission package, including the mandatory 8-page limit (which must account for a dedicated Limitations section), adherence to double-anonymous rules, and detailed sweeps for preserving author and lab anonymity in text, figures, and videos. Essential for ensuring compliance with complex conference guidelines.

This guide provides structured methodologies for authors to position their research within the context of existing literature and concurrent submissions at top-tier machine learning conferences (e.g., ICML, NeurIPS). It emphasizes identifying the technical delta, addressing novelty concerns, and properly citing related work to maximize acceptance chances.

A comprehensive guide for researchers submitting to major speech conferences (like INTERSPEECH). It details how to harden the reproducibility of speech results, covering crucial 'decay channels' such as corpus versions, text normalization rules (WER), objective metrics (EER), subjective protocols (MOS), and model stochasticity. The guide emphasizes pinning specific toolkits, commit hashes, configurations, and hardware details to ensure scientific rigor and verifiability.

A comprehensive guide for authors to determine the optimal conference venue for their speech-related research. It utilizes a 'modality test' to assess whether a contribution relies on audio signal characteristics (acoustics, prosody, etc.) or merely text transcriptions. The guide helps authors choose between major venues like INTERSPEECH, ACL, ICASSP, and specialized workshops, ensuring the claim is reviewed by relevant experts.

A comprehensive guide for researchers detailing how to establish rigorous provenance for on-device performance measurements (e.g., energy, latency, accuracy). It mandates recording all necessary context—including device model, SoC, OS build, framework versions, power instrument setups, and thermal conditions—to ensure results are verifiable and reproducible across different hardware or environments for academic publication.