数据与 AI技能

Run model training, fine-tuning, inference, and batch jobs on Modal’s cloud GPUs. Estimate hardware needs and cost, generate Python launch scripts, run or deploy workloads, monitor logs, and retrieve results without managing remote servers.

中文简介正在准备中。当前显示英语简介。

• Test CUDA code without a local GPU
• Run model training or LoRA fine-tuning on cloud GPUs
• Deploy a model inference API or vLLM service

Monitor model training through WandB metrics to detect invalid numbers, worsening loss, and declining evaluation results. Ambiguous trends receive a Codex review, while stopped runs retain key metrics and the reason for stopping.

中文简介正在准备中。当前显示英语简介。

• Check loss and evaluation trends during model training
• Detect NaN values and diverging loss
• Review flat or noisy training metrics

Launch machine learning experiments on local GPUs, remote servers, Vast.ai, or Modal. It checks the configured environment, deploys code, saves logs, and verifies startup, with optional W&B metric logging and automatic Vast.ai result collection and instance cleanup.

中文简介正在准备中。当前显示英语简介。

• Run training jobs on a local GPU
• Deploy experiments to an SSH server
• Run experiments on Vast.ai and collect results

Turn a research experiment plan into training and evaluation code, run a small setup check, and launch GPU experiments. Collect initial metrics and update progress records so the results are ready for further review.

中文简介正在准备中。当前显示英语简介。

• Implementing experiments from a written plan
• Checking a small run before full GPU experiments
• Collecting baseline and main-method results

Compare machine learning experiment results from JSON or CSV files. Calculate improvements over a baseline, summarize variation across repeated runs, identify trends and unusual results, and suggest experiments to test possible explanations.

中文简介正在准备中。当前显示英语简介。

• Compare model performance across experimental conditions
• Check variation across runs with different random seeds
• Identify trends as a parameter changes

Run batches of machine learning experiments on remote GPUs over SSH, varying settings and random seeds. The skill schedules jobs, tracks progress, retries GPU memory failures, and cleans up leftover sessions before starting subsequent batches.

中文简介正在准备中。当前显示英语简介。

• Run experiments across multiple random seeds
• Test combinations of model settings
• Chain teacher training and student distillation

Create consistent paper figures and comparison tables from experiment data. The skill selects suitable chart types, saves reusable plotting scripts, prepares LaTeX insertion code, and checks rendered figures for readability and comparisons for consistency with the underlying data.

中文简介正在准备中。当前显示英语简介。

• Plot training curves from experiment results
• Create method comparison and ablation tables
• Combine results across datasets into multi-panel figures

Check running experiments, collect logs and result metrics, and summarize progress alongside baseline comparisons. It supports SSH servers, Vast.ai, and Modal, with optional Weights & Biases metrics and cost reporting for the supported compute services.

中文简介正在准备中。当前显示英语简介。

• Check whether remote experiments have finished
• Compare collected metrics with the matching baseline
• Investigate NaN values or diverging training runs

Build an analytics web app from a Wren project's data definitions and business context. Create charts and tables, validate the app, and deploy to Vercel or Cloudflare with a checked URL for sharing.

中文简介正在准备中。当前显示英语简介。

• Building a sales dashboard from a Wren project
• Sharing a report with bundled data
• Creating an analytics app that fetches current data

Add business knowledge missing from a Wren project's database definitions. Compare source documents with existing context, then add explanations, rules, SQL examples, and proposed metrics through guided questions or an automatic mode with an audit report.

中文简介正在准备中。当前显示英语简介。

• Documenting status codes, units, and NULL meanings
• Adding business knowledge from handbooks and glossaries
• Defining aggregation metrics such as ARR or DAU

Find the Wren CLI workflow for answering data questions, connecting databases or SaaS sources, generating model definitions, adding business metrics, or building a shareable analytics app. Guides are retrieved from the installed CLI version.

中文简介正在准备中。当前显示英语简介。

• Set up Wren and connect a database
• Load HubSpot or Stripe data through dlt
• Generate model definitions and add business metrics

Create a Wren MDL project, a model layer for database queries, from an existing database. Discover tables and relationships, normalize column types, generate YAML files, and validate and build the project before indexing its schema.

中文简介正在准备中。当前显示英语简介。

• Create a Wren project from an existing database
• Add selected schemas and tables as a new data source
• Define model relationships from database constraints

Import SaaS data into DuckDB using dlt, a data loading tool, then generate a Wren project with table definitions and relationships. Verify the setup with real SQL queries, or start from an existing DuckDB file.

中文简介正在准备中。当前显示英语简介。

• Importing HubSpot or Stripe data for SQL analysis
• Creating a Wren project from an existing dlt DuckDB file
• Generating table definitions and parent-child relationships

Set up Wren Engine through guided steps covering environment checks, installation, database connection, and project creation. Prepare data definitions and validate the setup, then suggest initial questions to help the user begin exploring their data.

中文简介正在准备中。当前显示英语简介。

• Installing Wren Engine for the first time
• Connecting a new database to Wren
• Creating a project and preparing data definitions

Answer business data questions with Wren CLI by checking model definitions, reusing past queries, and executing SQL or predefined aggregations. The workflow also covers query verification, error diagnosis, and saving successful queries for future use.

中文简介正在准备中。当前显示英语简介。

• Analyze monthly revenue trends
• Rank customers by purchase value
• Compare metrics across plans or regions

Supports analysis of tables already in the workspace through data inspection, Python calculations, and charts chosen for the question. It also explains how to ask users for clarification and label results with meaningful units and names.

中文简介正在准备中。当前显示英语简介。

• Inspecting table columns, statistics, and sample rows
• Comparing categories or tracking changes over time
• Exploring distributions and relationships

Find missing data in connected sources, inspect table structure, and suggest loading options for analysis. It also helps prepare a connection form for a new source, leaving connection and loading choices to the user.

中文简介正在准备中。当前显示英语简介。

• Find external tables needed for an analysis
• Browse available connected data sources
• Check table columns before choosing data

Use Memori Cloud from Claude Code to consult stored preferences, decisions, and project history before substantive work, then save the exchange afterward. Recall can span sessions and use filters for dates and information types.

中文简介正在准备中。当前显示英语简介。

• Check previous decisions before coding
• Recall preferences and project requirements
• Review progress after time away

Use Memori through MCP, a protocol connecting agents to tools, to retrieve past decisions and preferences, summarize ongoing work, and resume after conversation compression. It also guides selective long-term memory storage and usage limit checks.

中文简介正在准备中。当前显示英语简介。

• Retrieve decisions and preferences from earlier sessions
• Review current work when starting a session
• Resume work after conversation context is compressed

Turn OpenClaw conversations, actions, and tool history into structured long-term memory. Agents can retrieve past facts, decisions, and summaries when needed, helping them continue work across sessions and recover context after it is shortened.

中文简介正在准备中。当前显示英语简介。

• Retrieve decisions made in previous sessions
• Review summaries of earlier work
• Recover context after conversation compaction

Run Customgpt operations through Rube MCP, which connects your agent to external tools. The skill searches for available actions, checks connection status, and follows current input requirements when executing supported tasks.

中文简介正在准备中。当前显示英语简介。

• Explore available Customgpt operations
• Check the Customgpt connection before execution
• Retrieve further pages of operation results

Run Docsbot AI operations through Rube MCP, a connection for agent tools. The skill provides a workflow for discovering supported actions, verifying an active connection, and executing operations with current tool inputs, including bulk processing guidance.

中文简介正在准备中。当前显示英语简介。

• Discover tools available for a Docsbot AI task
• Check authentication before running a task
• Process multiple operations through the available tools

Automate College Football Data operations through Rube MCP and Composio. Find tools for a requested task, check their input requirements and connection status, then execute operations individually or in batches.

中文简介正在准备中。当前显示英语简介。

• Find tools for a College Football Data task
• Verify the connection before running an operation
• Process multiple operations in a batch

Automate Apipie AI tasks through Rube MCP, a connection for agent tools. The skill discovers available operations, checks the service connection, and uses current input requirements to carry out supported workflows.

中文简介正在准备中。当前显示英语简介。

• Find tools for an Apipie AI task
• Check the Apipie AI connection before execution
• Run supported Apipie AI operations in bulk