数据与 AI技能

Helps build AI agents in .NET with persistent conversation threads. Covers messages, execution status, streaming responses, document search, and custom function calls using Azure.AI.Agents.Persistent, with examples for creating agents and cleaning up resources.

中文简介正在准备中。当前显示英语简介。

• Build an AI assistant that retains conversation threads
• Answer questions using uploaded documents
• Connect an agent to custom functions

Build Python text analysis with Azure AI Language to detect sentiment, names, key phrases, languages, and personal information. Examples also cover healthcare terminology extraction, combined analyses, authentication, and handling results for multiple documents.

中文简介正在准备中。当前显示英语简介。

• Analyzing sentiment about specific aspects of a review
• Detecting and redacting personal information
• Extracting names and key phrases from text

Create and manage persistent Azure AI agents from Java. Covers conversation threads that retain messages, agent execution, status checks, response retrieval, and cleanup, using preview SDK examples with guidance on authentication and cloud resource handling.

中文简介正在准备中。当前显示英语简介。

• Building a Java app with reusable AI agents
• Managing conversation threads and messages
• Checking execution states and handling timeouts

Helps implement document processing in TypeScript with Azure, extracting text, tables, invoice fields, and receipt details. Covers ready-made models, custom model training, document classification, authentication, and waiting for analysis results.

中文简介正在准备中。当前显示英语简介。

• Extracting text and tables from PDFs
• Reading invoice and receipt fields
• Training models for specialized documents

Build Java applications that analyze images with Azure AI Vision. Covers image descriptions, OCR for reading text, object and person detection, content tags, and suggested crop regions, with examples for authentication and handling results.

中文简介正在准备中。当前显示英语简介。

• Adding image descriptions to a Java application
• Extracting text from document images
• Detecting objects and people with their locations

Build Python image analysis with Azure AI Vision to generate captions and tags, locate objects and people, and extract text using OCR, or optical character recognition. It also provides suggested crop regions.

中文简介正在准备中。当前显示英语简介。

• Generate image captions and content tags
• Extract text from images
• Locate objects and people in photos

Implement .NET workflows for Azure AI Foundry projects, including agent creation, resource connections, dataset uploads, model deployment inspection, search index management, and evaluations. Examples also cover authentication, asynchronous operations, and resource cleanup.

中文简介正在准备中。当前显示英语简介。

• Creating and running agents from a .NET application
• Inspecting resource connections and model deployments
• Uploading datasets and managing search indexes

Build Python AI applications on Microsoft Foundry with azure-ai-projects. Provides examples for creating agents, connecting tools, managing conversations and project resources, and running evaluations through Foundry and OpenAI-compatible clients.

中文简介正在准备中。当前显示英语简介。

• Build an agent that searches uploaded documents
• Implement conversation threads and response handling
• List model deployments and project connections

Use Python to analyze astronomical data, convert celestial coordinates, units, and times, and read or write FITS astronomy files. It also guides catalog matching and calculations of cosmic distances and ages using cosmological models.

中文简介正在准备中。当前显示英语简介。

• Converting celestial coordinates for an observing location
• Reading FITS images and inspecting their headers
• Matching objects across astronomical catalogs

Design autonomous coding agents with patterns inspired by Cline and OpenAI Codex. Explore tool calling interfaces, permission and approval systems, browser automation, and workflows where people review or guide the agent's actions.

中文简介正在准备中。当前显示英语简介。

• Planning an autonomous coding agent
• Designing tool and function calling interfaces
• Defining permissions and approval steps

Design AI agents that break down goals, plan actions, use tools, and correct their own mistakes. The guidance focuses on reliable execution, clear operating boundaries, and managing conversation context as tasks grow longer.

中文简介正在准备中。当前显示英语简介。

• Designing an agent's planning and execution loop
• Breaking a complex goal into manageable tasks
• Reviewing self-correction and reliability patterns

Build vector search that considers both semantic similarity and patterns in how items connect. Use graph features to explore structure in embeddings, numerical representations of content, and compare retrieval results with conventional similarity search.

中文简介正在准备中。当前显示英语简介。

• Investigate structure missed by cosine similarity
• Include contextual roles in RAG retrieval
• Compare graph-based and cosine search results

Collect social posts, business listings, reviews, and other web data by selecting from over 55 Apify Actors, ready-made extraction tools. Run the selected tool, summarize results, and export data as CSV or JSON.

中文简介正在准备中。当前显示英语简介。

• Building prospect lists from business listings
• Collecting competitor posts and ads
• Gathering reviews for analysis

Automate Amplitude tasks through Rube MCP and Composio. Send tracking events, retrieve user activity, update user properties, and manage cohorts, saved groups of users, with checks for identifiers and asynchronous update status.

中文简介正在准备中。当前显示英语简介。

• Send page view and purchase events
• Retrieve a specific user's event history
• Update user plans and other properties

Help an AI coding assistant assess claims independently instead of automatically agreeing with the user. Responses lead with a conclusion and supporting evidence, while disagreements are evaluated for new information that could justify changing the assessment.

中文简介正在准备中。当前显示英语简介。

• Review claims without automatic agreement
• Ask for conclusions supported by evidence
• Reassess a position when new evidence appears

Retrieve financial data for market and company analysis, including stocks, forex, crypto, financial statements, and economic indicators. Access over 20 years of historical data and more than 50 technical indicators, which summarize price and trading patterns.

中文简介正在准备中。当前显示英语简介。

• Collect historical stock prices for analysis
• Compare company financial statements
• Track forex and crypto prices

Plan product analytics with event tracking, conversion funnels that reveal drop-offs, and retention analysis by user group. Includes PostHog examples and guidance for evaluating A/B tests, defining a core product metric, and designing dashboards.

中文简介正在准备中。当前显示英语简介。

• Designing events to track product usage
• Investigating drop-offs before signup or purchase
• Comparing weekly retention across user groups

Build Apache Airflow data pipelines with clear task dependencies and schedules. This skill guides workflow design, retries, monitoring, local testing, and production deployment, with patterns for investigating failed runs.

中文简介正在准备中。当前显示英语简介。

• Designing data pipelines and task dependencies
• Scheduling recurring batch jobs
• Testing Airflow workflows locally

Assess claims when a request pushes for agreement, invokes authority, or creates artificial urgency. The skill guides responses that separate evidence from persuasion, present strong counter-evidence first, and clearly state what remains uncertain.

中文简介正在准备中。当前显示英语简介。

• Evaluating claims under deadline pressure
• Reviewing requests that invoke authority
• Responding to demands to endorse unsupported claims

Plan AI application work across six areas: prompt evaluation, context allocation, retrieval-based answer generation, agent security, evaluation pipelines, and product strategy. This overview explains the approaches; the individual workflow packages, test cases, and CI templates are not included.

中文简介正在准备中。当前显示英语简介。

• Planning prompt evaluation before deployment
• Allocating context capacity and designing retrieval-based answers
• Planning an authorized AI agent security assessment

Design AI agents that use tools, plan tasks, retain relevant context, and recover from failures. Choose execution patterns, permission boundaries, memory handling, and runtime limits, and assess whether coordinating multiple agents is justified for the task.

中文简介正在准备中。当前显示英语简介。

• Choosing an execution loop for a tool-using agent
• Designing memory and recovery for long-running tasks
• Setting tool permissions, runtime limits, and cost caps

Improve workflows involving multiple AI agents by measuring bottlenecks, adjusting workload distribution, and managing model costs. This design guide helps compare changes against a measured baseline while checking quality, latency, and resource use.

中文简介正在准备中。当前显示英语简介。

• Finding bottlenecks in workflows involving multiple agents
• Reviewing task distribution and parallel execution
• Reducing model costs within quality thresholds

Create consistent records for evaluating AI agents, including task success, invalid tool calls, token use, and actual costs. Produce CSV tables, database definitions, and JSON validation schemas to support reproducible comparisons and checks for performance declines.

中文简介正在准备中。当前显示英语简介。

• Recording agent evaluation results in a shared format
• Comparing a candidate agent with a baseline
• Tracking actual cost per solved task

Convert human-written CLAUDE.md instructions into a structured label format for AI agents. Use it to review lengthy rules, reduce instruction overhead, and prepare prompts for migration between AI tools while preserving the intended rules.

中文简介正在准备中。当前显示英语简介。

• Restructuring a lengthy CLAUDE.md
• Reviewing instruction token usage
• Preparing rules for migration between AI tools