Skill UI

Guides product teams through hypothesis writing, metric definition, sample-size planning, ICE prioritization, and statistical interpretation so experiments stay practical, defensible, and aligned with business decision thresholds.

This skill explains complex concepts using the Socratic method. Instead of delivering direct answers or textbook definitions, it engages the user in a reflective, conversational dialogue. By guiding the user through thought experiments, small observations, and simple analogies, the assistant helps the user construct understanding step-by-step, facilitating deeper clarity through active reasoning. Ideal for understanding theories, abstract ideas, or system mechanics.

A structured methodology for de-risking product ideas and validating market opportunities. This framework guides teams through defining measurable outcomes, building Opportunity Solution Trees (OST), mapping critical assumptions (Desirability, Viability, Feasibility), and planning rigorous experiments (prototyping, interviews, usability tests) before committing to full development resources. Ideal for discovery sprints and pre-build validation.

Autonomous experiment loop that optimizes any file by a measurable metric. The agent edits a target file, runs a fixed evaluation, keeps improvements (git commit), discards failures (git reset), and loops indefinitely. Ideal for optimizing code speed, reducing bundle size, improving test pass rates, enhancing prompts, or boosting content quality through iterative measured attempts.

This skill manages and schedules recurring, autonomous experiments. It allows users to set an experiment to run continuously at predefined intervals (e.g., every 10 minutes, daily, or weekly) using cron scheduling logic. It handles the full lifecycle, including execution, result review, strategy refinement, and state persistence.

Allows users to resume an interrupted or paused autoresearch experiment. It automatically loads the full context, including configuration, strategy files, complete results history, and recent code changes. The system then summarizes the current best metrics and patterns, prompting the user to continue with a single iteration, set up an automated loop, or manually review the results.

This skill initializes a new autoresearch experiment. It guides the user through defining all necessary parameters, including the domain, target files, evaluation commands, metrics, and optimization direction. It is the essential starting point for automated optimization and performance benchmarking cycles.

This skill orchestrates end-to-end autonomous AI research projects using a sophisticated two-loop architecture. It manages the entire research lifecycle, from initial literature review and hypothesis formation to running constrained experiments and synthesizing findings into publishable papers. It acts as a project manager, autonomously driving the process while structuring the workspace and presenting progress to human users.

This tool analyzes ML experiment results from various paths (JSON/CSV). It computes comprehensive statistics, builds detailed comparison tables based on model types and hyperparameters, and generates actionable insights. Use it when you need to interpret raw experimental data, compare different model runs, or determine the next steps for research.

This skill transforms a refined research proposal or method idea into a detailed, structured, and claim-driven experiment roadmap. It helps users determine exactly which experiments to run, in what order, and what evidence is needed to support the core claims of a research paper, ensuring focus on novelty, simplicity, and genuine contribution.

A robust internal utility for integrating system status updates with Feishu/Lark. It allows other skills to send structured notifications (e.g., experiment complete, checkpoint waiting, review scores) via webhooks. Supports both 'Push only' (fire-and-forget) and 'Interactive' (bidirectional polling) modes, guaranteeing zero side effects if configuration is missing.

This skill generates and ranks publishable research ideas based on a broad direction provided by the user. It systematically validates concepts through pilot experiments and model reviews, optimizing for GPU efficiency and novelty. Suitable for brainstorming sessions, academic exploration, and identifying workable research paths.