Enable agentic autofix for code scanning alerts
Turn on Copilot's automated code fix suggestions for security alerts.
From GitHub Changelog: Copilot
Playbook for Copilot
303 prompts, settings, workflows and checklists seen with Copilot. News and docs live under Microsoft.
Coding127Agents and automation117Planning and PM114Writing62Learning and teaching59Analysis and data48Editing and humanizing43Research23
More for Copilot
Just added
- Enable agentic autofix for code scanning alertsTurn on Copilot's automated code fix suggestions for security alerts. · GitHub Changelog: Copilot
- GitHub Agentic autofix for code scanning alertsAutomated AI-powered remediation of code scanning security alerts. · GitHub Changelog: Copilot
- Use source material for context-aware AI editsProvide existing documents or data for AI to build or revise presentations. · OpenAI Help Center
- Enable HydraFusion and HydraFusion Max in CopilotAccess advanced AI model orchestration in VS Code, GitHub Copilot app, and CLI. · GitHub Changelog: Copilot
All 303
303 entries
GitHub Agentic autofix for code scanning alerts
Automated AI-powered remediation of code scanning security alerts.
From GitHub Changelog: Copilot
Reviewing ChatGPT outputs in PowerPoint
Key checks before relying on AI-generated or edited presentation content.
From OpenAI Help Center
Use source material for context-aware AI edits
Provide existing documents or data for AI to build or revise presentations.
From OpenAI Help Center
Add a specific slide with style matching
Instruct ChatGPT to add a slide after a specific section, matching the deck's style.
From OpenAI Help Center
Outline proposed slide changes before editing
Ask ChatGPT to first outline which slides it plans to change and why.
From OpenAI Help Center
ChatGPT for PowerPoint add-in for AI presentation assistance
AI sidebar in PowerPoint to create, edit, understand, and polish presentations.
From OpenAI Help Center
Enable HydraFusion and HydraFusion Max in Copilot
Access advanced AI model orchestration in VS Code, GitHub Copilot app, and CLI.
From GitHub Changelog: Copilot
Steps to build and deploy an AI agent from terminal
Checklist for planning, building, testing, and deploying an AI agent using GitHub Copilot CLI and Microsoft Foundry.
From Microsoft Developer on YouTube
Ship an AI agent from terminal using Copilot CLI and Foundry
Plan, build, test, and deploy a hosted AI agent from your terminal using GitHub Copilot CLI and Microsoft Foundry.
From Microsoft Developer on YouTube
Choosing the right AI model for specific security tasks
Don't always use the latest AI model; select models optimized for specific security tasks to improve efficiency and effectiveness.
From OpenAI News
Configure Copilot skills for specific procedures
Organize instructions and scripts in .github/skills folders for Copilot to load when relevant.
From GitHub Blog: Copilot
Define Copilot agent for recurring, multi-file tasks
Create custom agents in .github/agents for specialized roles in recurring, complex tasks.
From GitHub Blog: Copilot
Create path-specific Copilot instructions for relevant guidance
Use .github/instructions files with 'applyTo' to tailor Copilot guidance to specific file types.
From GitHub Blog: Copilot
Define repository-wide Copilot instructions for project standards
Add custom instructions to guide Copilot on project architecture, naming, and testing.
From GitHub Blog: Copilot
Repair code issues with Copilot inline chat
Use the new Fix action in diagnostic intention menus to have Copilot repair code issues.
From GitHub Changelog: Copilot
Control Copilot agent model for new conversations
Enterprise admins can set the default Copilot agent model for new chats in JetBrains IDEs.
From GitHub Changelog: Copilot
Letting Models Write Evaluation Criteria
Avoid having AI models generate your evaluation criteria for AI agent performance.
From Microsoft Source
Don't Wait for Models to Improve for Your Tech
Actively improve AI agent performance by controlling their information sources.
From Microsoft Source
Relying solely on AI model improvements for technology adoption
Don't wait for AI models to improve; actively shape the sources agents use to interact with your technology.
From Microsoft for Developers
AX Practitioner Playbook Evaluation Quality Checklist
Review your AI agent evaluation process before running it to ensure its quality and trustworthiness.
From Microsoft for Developers, Microsoft Source
AI agent uses wrong SDK version or deprecated pattern
Fix AI agent code errors by correcting the underlying documentation or training data sources.
From Microsoft for Developers
Debug slow or wrong AI agent responses using tracing
Connect Application Insights to your Microsoft Foundry project for server-side tracing to pinpoint agent issues.
From Microsoft Developer on YouTube
Use Claude Haiku 5.5 for fast coding tasks in GitHub Copilot
Claude Haiku 5.5 is now available in GitHub Copilot for fast, high-volume coding tasks.
From GitHub Changelog: Copilot
Enable Local Sandboxing for GitHub Copilot Agentic Workflows
Configure GitHub Copilot to use local sandboxes for agentic workflows, restricting tool execution access to system resources.
From GitHub Changelog: Copilot
GitHub Copilot Local Sandboxing for Secure Agentic Workflows
Provides a secure execution boundary for agentic workflows on your machine, restricting access to filesystem, network, and credentials.
From GitHub Changelog: Copilot
Troubleshoot GitHub Copilot CLI provider connection failures
Identify and resolve GitHub Copilot CLI provider connection failures using error messages in the model picker.
From GitHub Changelog: Copilot
Configure GitHub Copilot app to use local model providers
Add supported local model providers in the GitHub Copilot app settings to choose models that fit your work.
From GitHub Changelog: Copilot
Discover and add local Ollama models in GitHub Copilot CLI
Use the /model command to find and add local Ollama models to GitHub Copilot CLI without leaving your workflow.
From GitHub Changelog: Copilot
Enable explicit offline mode in GitHub Copilot CLI
Set the COPILOT_OFFLINE=true environment variable to enable explicit offline mode in GitHub Copilot CLI.
From GitHub Changelog: Copilot
Convert meeting transcripts to project artifacts using GitHub Copilot agents
Automate project plan generation from meeting transcripts and documents with custom Copilot agents.
From Microsoft Developer on YouTube
Choosing an AI Code Reviewer Based on Workflow Needs
Select an AI code reviewer by understanding its strengths in catching different types of issues and its precision-recall trade-offs.
From GitHub Blog: Copilot
Azure Canvases for GitHub Copilot
Interactive workspaces for Azure resource queries, cost checks, and Azure Functions skills.
From GitHub Changelog: Copilot
Describe Outcome, Apps, and Constraints for Copilot Computer Use
Provide clear instructions including desired outcome, applications, and constraints for effective Copilot automation.
From GitHub Changelog: Copilot
GitHub Copilot Computer Use for Desktop App Interaction
GitHub Copilot can now control desktop apps for automation tasks, even without APIs.
From GitHub Changelog: Copilot
Attach GitHub context to Copilot chats
Add GitHub issues or pull requests to Copilot chats for relevant context.
From GitHub Changelog: Copilot
Authoring dynamic workflows with Copilot assistance
Leverage Copilot's built-in guidance to learn about or fully generate dynamic workflows.
From GitHub Changelog: Copilot
When to use dynamic workflows versus standard chat prompts
Use dynamic workflows for reusable, staged processes; use standard chat for quick, simple tasks.
From GitHub Changelog: Copilot
Copilot dynamic workflows for complex agent orchestration
Define multi-agent tasks in code for reliability and observability in complex scenarios.
From GitHub Changelog: Copilot
Intermountain Health uses Dragon Copilot for automated clinical note drafting
Automating clinical note drafting with AI to reduce documentation time and improve patient focus.
From Microsoft Source
KARL STORZ uses Microsoft 365 Copilot for meeting summaries and drafting
AI tool enhances meeting productivity and information retrieval for MedTech organization.
From Microsoft Source
Pierre Fabre uses generative AI for document summarization and content generation
Generative AI platform accelerates scientific discovery and daily tasks for employees.
From Microsoft Source
Use GPT-6.1 Sol for Agentic Coding and Terminal Workflows
GPT-6.1 Sol offers strong multistep coding and efficient token use for agentic and terminal tasks.
From GitHub Changelog: Copilot
Manage GPT-6.1 Sol Access for Copilot Business/Enterprise
Administrators control GPT-6.1 Sol availability via Copilot settings.
From GitHub Changelog: Copilot
Select GPT-6.1 Sol in GitHub Copilot Model Picker
Choose GPT-6.1 Sol for advanced coding and terminal tasks within Copilot.
From GitHub Changelog: Copilot
Remap Copilot key on Windows 11 using PowerToys
Remap the dedicated Copilot key on your Windows 11 PC to a different function using PowerToys.
From Windows Central
Remap Copilot key in Windows 11 settings
Customize the Copilot key's function directly within Windows 11 settings.
From Windows Central
Reviewing AI agent code changes with GitHub Copilot app
Steps to confirm AI-generated code fixes bugs and meets requirements before merging.
From Microsoft for Developers
Manage Claude Sonnet 5.5 access for Copilot Business/Enterprise
Admins can manage Claude Sonnet 5.5 access via the model policy in Copilot settings for Business and Enterprise plans.
From GitHub Changelog: Copilot, GitHub Changelog: Copilot
Select Claude Sonnet 5.5 in GitHub Copilot's model picker
Users can choose Claude Sonnet 5.5 within GitHub Copilot's model picker for Copilot Pro, Pro+, Max, Business, and Enterprise plans.
From GitHub Changelog: Copilot
GitHub Copilot enterprise managed settings in-product validator
Validate GitHub Copilot enterprise managed settings directly within the product interface.
From GitHub Changelog: Copilot
AI agent disclosed other customer data; reduced disclosure rate
Reduced customer data disclosure by implementing runtime policy after risk discovery.
From Microsoft Source
run-assert-eval skill for AI agent risk discovery and fixes
Discovers AI agent risks, measures failures, generates runtime policy, and proves fixes.
From Microsoft Source
Automated AI agent risk discovery, policy generation, and validation
Automates AI agent risk discovery, policy creation, and fix validation using run-assert-eval.
From Microsoft Source
Create Custom Workflows with Plain English Descriptions
Use natural language to define and generate custom interfaces for AI agents.
From GitHub Blog: Copilot
Refine Existing Canvas Extensions
Customize pre-built canvas extensions to fit your specific workflow needs.
From GitHub Blog: Copilot
GitHub Copilot App Canvases for Custom Workflows
Build custom, shareable interfaces for your workflows without coding using plain English descriptions.
From GitHub Blog: Copilot
Enable Copilot Memory for Agentic Autofix
Activate Copilot Memory to provide context and learning for automated security fixes.
From GitHub Changelog: Copilot
Agentic autofix with Copilot Memory
Automated security fix generation that uses past fixes for context and learns new patterns.
From GitHub Changelog: Copilot
Microsoft Copilot Studio for AI agent building
Build custom AI copilots and agents with a low-code interface.
From Microsoft Developer on YouTube, Microsoft Developer on YouTube and 1 more
Evaluate AI model performance and cost
Checklist for comparing AI models on key metrics like latency, tokens, quality, and cost.
From Microsoft Developer on YouTube
Pick AI models using data not intuition
Use a model catalog to compare AI models on cost, latency, and quality for informed selection.
From Microsoft Developer on YouTube
Configure OpenTelemetry for GitHub Copilot Agents
Enable and configure OpenTelemetry for GitHub Copilot agent activity monitoring via enterprise settings.
From GitHub Changelog: Copilot
Use Claude Opus 5.5 in GitHub Copilot for coding tasks
Claude Opus 5.5 is now available in GitHub Copilot for agentic coding and knowledge work.
From GitHub Changelog: Copilot
Copilot Impact Dashboard for Enterprise Administrators
Monitor Copilot feature adoption and user engagement for enterprise developers.
From GitHub Changelog: Copilot
Customize Agents from GitHub Copilot Chat Window
Build and customize agents directly within the GitHub Copilot Chat window using Foundry Canvas.
From Microsoft Developer on YouTube
Applying AI to existing, inefficient workflows
Don't speed up broken processes; redesign workflows end-to-end before adding AI agents.
From Microsoft Source
Applying AI to existing steps without workflow redesign
Do not apply AI to broken processes; redesign workflows end-to-end first for true transformation.
From The Official Microsoft Blog
Treating AI like a traditional technology rollout
Avoid deploying AI tools with basic training and expecting transformation; focus on business goals and user workflows instead.
From The Official Microsoft Blog
Approve Copilot budget increase requests in GitHub
Manage Copilot credit requests directly within GitHub organization or enterprise settings.
From GitHub Changelog: Copilot
Prepare for Microsoft Copilot PowerPoint webinar Q&A
Questions to ask experts about using Copilot and Brand Kit in PowerPoint.
From Microsoft 365 on YouTube
Configure Copilot custom property suggestions policy
Control availability of Copilot's custom property suggestions for your organization.
From GitHub Changelog: Copilot
Checklist for production-ready AI agent refactoring
Steps to ensure an AI agent is production-ready after refactoring.
From Microsoft Developer on YouTube, Microsoft Developer on YouTube
Refactor an AI agent using GitHub Copilot and Foundry skills
Use GitHub Copilot with Microsoft Foundry skills to review and refactor an existing AI agent.
From Microsoft Developer on YouTube
Configure Copilot auto model selection tiers
Choose efficiency, balance, or intelligence to balance cost, quality, and response time.
From GitHub Changelog: Copilot
Define AGENTS.md for Copilot to follow runbooks and automate tasks
Use AGENTS.md to write runbooks for Copilot, defining how to name campaigns, map dates, and format emails.
From GitHub Blog: Copilot
Relying on rigid automation instead of conversational AI
Avoid overly rigid automation; use conversational AI to allow for flexibility and human input in workflows.
From GitHub Blog: Copilot
Automate repetitive work by writing down runbooks for AI
If you can document your work process, you can automate it using AI tools like GitHub Copilot.
From GitHub Blog: Copilot
Use GitHub Issues as the central unit of work for marketing projects
Treat GitHub Issues as the primary workspace for marketing projects, integrating planning, discussion, and status updates.
From GitHub Blog: Copilot
GitHub Copilot for automating marketing operations
GitHub Copilot can automate repetitive marketing tasks by interpreting documented runbooks and engaging in conversational workflows.
From GitHub Blog: Copilot
Automate event setup from a single GitHub Issue using Copilot and Actions
Trigger event setup automation from a GitHub Issue, using Copilot for initial setup and GitHub Actions for subsequent tasks.
From GitHub Blog: Copilot
Check agent-generated code using diff, terminal, and browser panels
Review, run, and preview AI code changes side by side without leaving the Copilot app.
From GitHub Blog: Copilot
Use integrated panels for AI code review loop
Streamline AI code review by using the diff, terminal, and browser panels within the Copilot app.
From GitHub Blog: Copilot
Using Copilot app's browser panel for UI preview
Test user interfaces and new features directly in the browser panel within the Copilot app.
From GitHub Blog: Copilot
Using Copilot app's diff panel for code changes
View additions in green and deletions in red to understand AI code modifications.
From GitHub Blog: Copilot
Using Copilot app's terminal panel for running code
Execute project commands and read results directly within the Copilot app's terminal.
From GitHub Blog: Copilot
Embed safety and age-appropriateness from the start
Design AI experiences with built-in protections and age-specific features.
From Microsoft Source
Update GitHub Copilot workflows from MAI-Code-1-Flash to a supported model
Replace the deprecated MAI-Code-1-Flash model in GitHub Copilot with an alternative model to maintain functionality.
From GitHub Changelog: Copilot
Building interactive Copilot apps with Python, FastMCP, and Prefab
Guide on creating dynamic interfaces for AI clients using Python and specific libraries.
From Microsoft Developer on YouTube
FastMCP for building interactive Microsoft Copilot apps
Library for creating interactive interfaces within AI clients like Microsoft Copilot.
From Microsoft Developer on YouTube
Centrally control GitHub Copilot agent operations with managed permissions
Administer GitHub Copilot Business or Enterprise to centrally control agent operations like shell commands and file edits.
From GitHub Changelog: Copilot
Steps for building production-ready AI agents
Key considerations for developing AI agents that are ready for production use.
From Microsoft Developer on YouTube
Evaluate AI agent performance and token usage
Measure and review an AI agent's performance and token consumption using evaluation tools.
From Microsoft Developer on YouTube
Microsoft Foundry Portal for building AI agents
Build AI agents without code, deploy models, and connect servers for agent functionality.
From Microsoft Developer on YouTube
Build a Cupcake Ordering AI Agent with Foundry Portal
Step-by-step guide to creating, testing, and evaluating an AI agent for a cupcake store.
From Microsoft Developer on YouTube
Evaluate AI models on your specific tasks, not just public benchmarks
Test AI models using your own repositories and agent setup to get accurate performance results for your team's workflow.
From Microsoft for Developers
Choose the minimal viable model for your work, not the smartest
Select the least expensive AI model that meets your specific task requirements to optimize cost and performance.
From Microsoft for Developers
Azure Chaos Studio outage simulation and reporting
Steps to simulate failures, monitor them, and generate reports using Azure Chaos Studio.
From Microsoft Azure on YouTube
Azure Chaos Studio for simulating cloud application failures
Safely simulate failure conditions in Azure applications to find weaknesses before real outages occur.
From Microsoft Azure on YouTube
Simulate and report on Azure application resilience with Chaos Studio
Create, run, and monitor failure simulations in Azure Chaos Studio, then generate reports for review.
From Microsoft Azure on YouTube
Be specific and provide context in AI prompts
Clearly identify the task, provide necessary context, and set the preferred tone and style for AI responses.
From OpenAI Help Center
Break down complex AI tasks into smaller prompts
Divide complex requests into smaller, focused prompts for better AI outcomes and iterative refinement.
From OpenAI Help Center
Iterate and refine AI prompts through conversation
Treat prompting as a conversation, refining requests based on initial AI answers and experimenting.
From OpenAI Help Center
Treat AI prompts like requests to a human
Provide context and details in AI prompts as you would when asking a person to complete a task.
From OpenAI Help Center
Delegate larger tasks with GPT-6 Astra in Copilot
Leverage GPT-6 Astra to hand off complex tasks, focusing on review and decision-making.
From Microsoft 365 Copilot Blog
Ground GPT-6 Astra with Work IQ for relevance
Use Work IQ to connect GPT-6 Astra to your data for more accurate AI responses.
From Microsoft 365 Copilot Blog
Manage GPT-6 Astra access for Copilot users
Admins can manage GPT-6 Astra access via Copilot settings model policy.
From GitHub Changelog: Copilot
Select GPT-6 Astra model in Copilot
Users can choose the GPT-6 Astra model from the model picker in GitHub Copilot.
From GitHub Changelog: Copilot
Selectivity is key to multi-model orchestration
Choose the least complex workflow that meets quality needs, using more powerful models only when necessary.
From GitHub Blog: Copilot
HydraFusion selects best model for task
HydraFusion dynamically chooses the most suitable model or model combination for each specific coding task.
From GitHub Blog: Copilot
Access HydraFusion via GitHub Copilot CLI
Use /experimental in GitHub Copilot CLI to select HydraFusion as a model.
From GitHub Blog: Copilot
Project HydraFusion for GitHub Copilot
Orchestrates multiple AI models to draft, critique, and revise code for optimal quality and cost.
From GitHub Blog: Copilot
Improving code generation with execution context constraints
Providing AI models with execution constraints like RAM and time significantly improves code efficiency and success rate.
From arXiv cs.AI
AI deployment readiness checklist
Key considerations before deploying AI models into production.
From Microsoft Source
Configure safeguards for AI agent actions
Implement access controls, approvals, and monitoring for AI agents to manage risk.
From Microsoft Source
Choose Astra deployment: Standard vs. Provisioned Throughput
Select deployment options based on workload needs for latency and throughput.
From Microsoft Source
Integrate Codex CLI for agent-style coding in local environments
Use the open-source Codex CLI to run agent-style coding directly on local repositories for iterative review and edits.
From OpenAI Developers Blog
Prompt Codex to draft documentation scaffolds from code
Generate initial documentation drafts by having Codex read framework code in a monorepo.
From OpenAI Developers Blog
Use Codex to explain pull requests and code sections
Leverage Codex with the gh command to get summaries and explanations of code changes and features.
From OpenAI Developers Blog
Custom repository rules for Codex Code Review in AGENTS.md
Define custom rules in AGENTS.md to guide Codex Code Review for specific repository conventions and issues.
From OpenAI Developers Blog
Evaluating Codex Code Review rule effectiveness
Check if Codex surfaces intended violations, avoids false positives, catches bugs, and cites guidance correctly.
From OpenAI Developers Blog
Preserve experimental notification names for backward compatibility
Do not rename experimental notifications if consumers rely on them.
From OpenAI Developers Blog
Codex Code Review with custom repository rules
AI-powered code review that uses custom rules defined in AGENTS.md for better convention adherence.
From OpenAI Developers Blog
Configuration resolution for Claude M365 add-ins
Understanding the order of settings applied when Claude M365 add-ins load.
From Claude Help Center
LLM gateway request flow for Claude M365 add-ins
How requests are routed from the M365 add-in through your gateway to the LLM provider.
From Claude Help Center
Access Claude models via Python, TypeScript, C# SDKs in Foundry
Integrate Claude models into your Azure applications using provided SDKs with Microsoft Entra authentication.
From Claude Help Center
AI initiatives slowing down on infrastructure fundamentals
Allowing identity, networking, governance, and compliance to hinder AI project progress.
From Microsoft Azure Blog
Enterprise AI deployment and governance checklist
Key considerations for deploying and managing AI models in an enterprise setting.
From Microsoft Azure Blog
Optimize AI model consumption with token efficiency
Utilize models designed for token efficiency on complex work to manage consumption.
From Microsoft Azure Blog
Containment for AI agent computer-use capabilities
Implement controls to manage the risks associated with AI agents interacting with applications.
From Microsoft Azure Blog
Run multiple AI agents in parallel for increased productivity
Manage several independent AI agent sessions simultaneously to get more done in less time.
From GitHub Blog: Copilot
AI agent evaluation setup and execution checklist
Steps to set up and run evaluations for your AI agent.
From Microsoft Developer on YouTube
Microsoft Foundry Toolkit for building AI agents in VS Code
Build, test, and deploy AI agents directly within your VS Code environment.
From Microsoft Developer on YouTube
Build and test an AI agent using Microsoft Foundry Toolkit
Create, configure, test, and evaluate an AI agent within VS Code.
From Microsoft Developer on YouTube
Start agentic AI projects with the right use case
Focus on internal processes worth automating to ensure success for agentic AI solutions.
From Microsoft Azure on YouTube
Control effort level and cost in Microsoft Copilot's Cowork
Users can select effort levels (Light, Medium, Heavy) to balance quality, speed, and credit usage.
From Microsoft 365 Copilot Blog
IT admin capabilities for Copilot AI Credits and Data Export
IT admins can now manage GitHub Copilot AI Credits and export data related to Copilot usage.
From Microsoft 365 Copilot Blog
Copilot content exclusions protect sensitive code
Content exclusion policies in Copilot prevent sensitive code from being used as context in agentic workflows.
From GitHub Changelog: Copilot
Configure Copilot content exclusions for enterprise and orgs
Set up content exclusion policies in Copilot to prevent sensitive code from being used as context.
From GitHub Changelog: Copilot
Avoid computer-based training for AI adoption
Prefer small group peer learning over computer-based training for AI adoption.
From Microsoft Source
Kier Group uses Copilot prompt-a-thons to drive AI adoption
Kier Group used prompt-a-thons and peer learning to encourage AI adoption among employees.
From Microsoft Source
Ask Copilot to identify leadership strengths and blind spots
Use Copilot to analyze your work documents and identify personal leadership strengths and areas for growth.
From Microsoft Source
Define goals, metrics, and loop for AI progress
AI progress means defining goals, evaluation metrics, and iterative system improvement.
From Microsoft Developer on YouTube
Grounding AI models in user data for relevance
AI results improve when grounded in your specific files and business data.
From Microsoft 365 Copilot Blog
Manage GPT-6 Astra availability in Microsoft 365 admin center
Admins can control GPT-6 Astra access and configuration for their organization.
From Microsoft 365 Copilot Blog, Microsoft 365 Copilot Blog
Anthropic Claude Fable 5.1 in Microsoft Copilot
A new frontier model for complex tasks in Microsoft Copilot.
From Microsoft 365 Copilot Blog
Copilot approval does not count by default
Copilot's approval assessment alone does not satisfy merge requirements unless explicitly enabled.
From GitHub Changelog: Copilot
Copilot approval is dismissed on new commits
If new commits are pushed after Copilot approves, its approval is automatically dismissed.
From GitHub Changelog: Copilot
Student Developer Pack and Copilot setup
Steps to claim the GitHub Student Developer Pack and install Copilot.
From Microsoft Developer on YouTube
Controlling Copilot with different interaction modes
Utilize Interactive, Plan, and Autopilot modes for effective GitHub Copilot control.
From Microsoft Developer on YouTube
Extending Copilot with MCP servers and skills
Enhance Copilot's capabilities using built-in tools and custom agent skills.
From Microsoft Developer on YouTube
Prompting a resume PDF into a portfolio website
Use GitHub Copilot to transform a resume PDF into a live portfolio site without coding.
From Microsoft Developer on YouTube
GitHub Copilot for students building a portfolio
AI pair programmer to help students build a live portfolio website with no coding experience.
From Microsoft Developer on YouTube
Build a portfolio website using GitHub Copilot
Step-by-step process to create a live portfolio website from a resume PDF using AI.
From Microsoft Developer on YouTube
DevOps CI/CD Automation with AI Agents on Azure
Automating DevOps CI/CD issue detection, diagnosis, and rollback using AI agents.
From Microsoft Developer on YouTube
Standardize AI Agent Tool Integration for Production
Use standard protocols like MCP to ensure AI agents can reliably call tools in production.
From Microsoft Developer on YouTube
MCP for AI Agents to Discover and Invoke Tools
MCP provides a standard way for AI agents to discover and invoke tools at runtime.
From Microsoft Developer on YouTube
Top performing models for scientific claim validation
Claude Opus, Gemma, GPT-5.5, and Claude Fable showed strong performance in scientific claim verification tasks.
From arXiv cs.AI
Microsoft marketing scales product launches with AI agents
AI agents help Microsoft marketing manage 149% more product launches by automating consistency.
From Microsoft Azure on YouTube
Automate consistency, not judgment, with AI
Use AI for tasks requiring consistency, leaving strategic judgment to humans.
From Microsoft Azure on YouTube
AI-powered blog draft review workflow
Automate blog draft evaluation against expert-defined rubrics using AI agents.
From Microsoft Azure on YouTube
AI agent development and deployment checklist
Steps to follow for building and deploying an AI agent.
From Microsoft Developer on YouTube
Consider the full AI agent development cycle
Think about agent production from model connection to evaluation.
From Microsoft Developer on YouTube
WinGet configuration file for dev machine setup
Template for a WinGet configuration file to automate dev machine setup.
From Microsoft Developer on YouTube
WinGet configuration for dev machine setup
A WinGet configuration file to set up a development machine in one pass.
From Microsoft Developer on YouTube
Getting started with Azure SRE Agent trial
Steps to begin using the Azure SRE Agent trial for operational improvements.
From Microsoft for Developers
Utilize Azure SRE Agent 30-day trial for cost-free setup
New customers get a 30-day trial for Azure SRE Agent, waiving setup and ready-time charges.
From Microsoft for Developers
Start Azure SRE Agent trial with a familiar incident
Use a known incident type for your first Azure SRE Agent test to easily measure success.
From Microsoft for Developers
Azure SRE Agent Live Reports for dynamic operational views
Live Reports provide continuously updated operational views from connected systems.
From Microsoft for Developers
Azure SRE Agent VNet integration for private resource access
VNet integration allows the agent to access private resources without opening network boundaries.
From Microsoft for Developers
AI as a patient, interactive tutor for learning complex topics
Use AI to learn difficult subjects by asking questions and getting patient, interactive explanations.
From Simon Willison
Create custom views in the My work pane
Build personalized views in the My work pane to filter and organize your tasks.
From GitHub Blog: Copilot
Start Copilot sessions directly from My work items
Initiate new Copilot sessions with specific issues or pull requests as context.
From GitHub Blog: Copilot
Use the My work pane to track Copilot sessions
Organize and track all your Copilot requests and work in one centralized location.
From GitHub Blog: Copilot
Create a new issue directly from Copilot
Quickly create a new issue with Copilot, setting it as the context for a new session.
From GitHub Blog: Copilot
Filter repositories in the Copilot app
Use the repository filter to scope Copilot sessions to specific projects.
From GitHub Blog: Copilot
Switch between list and table views in My work
Choose between list and table views for different levels of control over how your work is displayed.
From GitHub Blog: Copilot
Relying solely on chat for complex agentic workflows
Chat becomes a long scroll of buried instructions, logs, and corrections, making work hard to track and audit.
From GitHub Blog: Copilot
Steps for building a minimal canvas around a repeated workflow
Start small with /create-canvas, run real work, and iterate based on actual usage.
From GitHub Blog: Copilot
Invest in workflow architecture for long-term AI cost savings
Use durable surfaces like canvases to reduce repeated prompting, context loss, and rework, saving time and money.
From GitHub Blog: Copilot
Java Modernization Studio canvas example
A template for structuring AI-driven Java modernization workflows with explicit phases and validation gates.
From GitHub Blog: Copilot
Site Studio canvas for personal content management
A template for managing personal site content with AI, focusing on section progress and iterative edits.
From GitHub Blog: Copilot
Carrying context across multiple disconnected tools
Avoid manually switching between tools and re-entering context for each step of the SDLC.
From GitHub Blog: Copilot
Pre-merge deployment risk assessment checklist
Questions to ask before merging code to ensure a safe deployment.
From GitHub Blog: Copilot
Integrate external tools into GitHub with agent apps
Bring Amplitude, Endor Labs, LaunchDarkly, and PagerDuty into GitHub to streamline workflows.
From GitHub Blog: Copilot
Centralize development context within GitHub
Bring your tools and their context into GitHub to reduce context switching.
From GitHub Blog: Copilot
Use agent apps to scope, secure, roll out, and ship a feature
Leverage GitHub agent apps to analyze data, check dependencies, set feature flags, and assess deploy risk.
From GitHub Blog: Copilot
Use Copilot for Instant Creative Sparks
Leverage Copilot to generate story elements and overcome writer's block quickly.
From Microsoft 365 on YouTube
Generate Story Elements with Copilot
Ask Copilot to create characters, settings, conflict, or plot twists to overcome writer's block.
From Microsoft 365 on YouTube
Start prompting GitHub Copilot with simple, plain language requests
Begin with a clear, simple description of the change you want, and refine as needed.
From GitHub Blog: Copilot
Choose AI model in GitHub Copilot app for different tasks
Select the AI model that best suits your task's complexity in the GitHub Copilot app.
From GitHub Blog: Copilot
Connect GitHub Copilot app to a repository or local folder
Connect your project to give GitHub Copilot access to your code and files for context.
From GitHub Blog: Copilot
Use voice input for prompts in GitHub Copilot app
Utilize voice input in GitHub Copilot to describe longer or complex requests naturally.
From GitHub Blog: Copilot
Building AI agents without defining the problem first
Avoid starting AI agent development with technology instead of a clear problem.
From Microsoft Source: AI
Steps for building an AI agent in Microsoft Copilot
A checklist for creating and refining AI agents using Microsoft Copilot.
From Microsoft Source: AI
Connect agents to relevant data sources
Link your AI agent to specific information sources for accurate output.
From Microsoft Source: AI
Refine agent instructions for specific behavior and tone
Shape agent responses by defining its behavior, tone, and specific tasks.
From Microsoft Source: AI
Start agent building with the problem, not the tech
Focus on the problem to solve before choosing or building an AI agent.
From Microsoft Source: AI
Build an AI agent using Microsoft Copilot
Guide to creating an AI agent in Microsoft Copilot without coding.
From Microsoft Source: AI
Slash commands enhance Copilot app interaction
Slash commands provide shortcuts for managing sessions, navigating projects, and customizing Copilot workflows.
From GitHub Blog: Copilot
Automate coding implementation with Copilot
Use /autopilot to let Copilot work through implementation steps towards a given goal.
From GitHub Blog: Copilot
Coordinate work across multiple Copilot sessions
Use /orchestrate to coordinate work across sessions and repositories for larger efforts.
From GitHub Blog: Copilot
Create interactive interfaces from Copilot chat
Use /create-canvas to turn conversations into visualizations, dashboards, or custom workflows.
From GitHub Blog: Copilot
Generate a plan for a coding task
Use /plan to break down a task, think through your approach, and identify next steps.
From GitHub Blog: Copilot
Get a second opinion on code with Copilot
Use /rubber-duck for an independent review of your work to surface blind spots.
From GitHub Blog: Copilot
Pressure-test a coding approach with Copilot
Use /spar to have Copilot question assumptions and point out risks or tradeoffs.
From GitHub Blog: Copilot
Use AI to quickly grasp presentation content
Leverage AI tools to get instant summaries of presentations, saving time on review.
From Microsoft 365 on YouTube
Microsoft 365 Copilot for Presentation Summarization
AI assistant that summarizes presentations, highlights key takeaways, decisions, and next steps.
From Microsoft 365 on YouTube
Adopt AI tools like Copilot within a 30-day timeframe
Focus on practical use cases and value evaluation for rapid AI adoption in SMBs.
From Microsoft 365 Copilot Blog
Waiting for perfect software instead of building custom AI tools
Don't wait for vendors; use tools like Copilot CLI to build AI solutions tailored to your needs.
From GitHub Blog: Copilot
GitHub legal team uses Copilot CLI to build contract drafting tool
Lawyers built a contract drafting tool using Copilot CLI, cutting review time in half.
From GitHub Blog: Copilot
Legal team automates DMCA notice analysis with Copilot CLI
A legal team used Copilot CLI to create a repeatable workflow for analyzing DMCA notices.
From GitHub Blog: Copilot
Build AI tools around your own judgment and experience
Use AI tools like Copilot CLI to build solutions that reflect your unique way of working and expertise.
From GitHub Blog: Copilot
Operationalize legal knowledge with plain language instructions
Embed legal standards and methodologies into AI workflows using readable Markdown and plain language.
From GitHub Blog: Copilot
Add agents to Copilot Chat by @mentioning
Add Word, Excel, and PowerPoint agents directly into Copilot Chat prompts using @mention.
From Microsoft 365 Copilot Blog
Use Copilot in Excel to track reading goals
Leverage Copilot in Excel to create a personalized reading tracker and dashboard for better goal management.
From Microsoft 365 on YouTube
Microsoft Copilot in Excel for reading trackers
Use Copilot in Excel to build a personalized reading dashboard and track book data visually.
From Microsoft 365 on YouTube
Treating AI adoption as a product launch
Avoid launching AI as a standalone product; integrate it into broader business transformation.
From Microsoft on YouTube
Build agents on Claude Opus 5 in Copilot Studio for advanced reasoning
Copilot Studio allows building agents using Claude Opus 5 for advanced reasoning and workflow automation.
From Microsoft 365 Copilot Blog
Generate Excel Pivot Table with Copilot
Prompt Copilot to create a pivot table from your data, summarizing key insights.
From Microsoft 365 on YouTube
Microsoft 365 Copilot for Excel Pivot Tables
AI assistant that builds Excel pivot tables from simple prompts, no formulas needed.
From Microsoft 365 on YouTube
Assuming LLM reasoning matches human intuition
Do not assume LLM reasoning aligns with human expectations or intuition.
From Microsoft for Developers
Shipping agent experience changes without local testing
Avoid shipping untested agent experience changes to production by using local emulation.
From Microsoft for Developers
Agent ignores documentation tips; use warnings for plan failure
Use warnings about plan failure, not tips, to change agent behavior.
From Microsoft for Developers
Make agent experimentation cheap for faster iteration
Reduce the cost of each experiment to increase the number of hypotheses tested.
From Microsoft for Developers
Treat agent improvements as hypotheses needing measurement
Every proposed agent improvement is a hypothesis that requires empirical testing.
From Microsoft for Developers
Dev Proxy for local agent request emulation
Emulate agent request responses locally to test changes without deploying.
From Microsoft for Developers
Azure Copilot migration agent workflow steps
Steps for using the Azure Copilot migration agent for storage migration.
From Microsoft Azure on YouTube
Azure Copilot migration agent for storage migrations
Automates and guides on-premises SMB file share migrations to Azure Files.
From Microsoft Azure on YouTube
Leverage Copilot for quick answers from past emails
Use Copilot in Outlook to quickly get answers from your past email correspondence.
From Microsoft 365 on YouTube
Refine and correct emails before sending with Copilot
Use Copilot in Outlook to correct and refine your email drafts before sending.
From Microsoft 365 on YouTube
Drafting professional emails with Copilot in Outlook
Use prompts to guide Copilot in Outlook to draft professional emails.
From Microsoft 365 on YouTube
Copilot in Outlook for email drafting and inbox assistance
Draft emails, check grammar, adjust tone, and get inbox context with Copilot in Outlook.
From Microsoft 365 on YouTube
Re-explaining context when switching applications
Avoid re-explaining context when moving work between applications; use tools that carry context automatically.
From Anthropic News
Contextual work across Microsoft 365 apps with Claude
Seamlessly transfer context and work between Excel, PowerPoint, Word, and Outlook using Claude add-ins.
From Anthropic News
Applying AI to broken workflows without redesign
Don't use AI to automate inefficient processes; fix the process first.
From Microsoft on YouTube
General availability for Copilot Cowork worldwide
Copilot Cowork is now available globally for users to automate task execution.
From Microsoft 365 Copilot Blog
Checklist for creating a newsletter with Copilot in Word
Steps to follow when using Copilot to generate and refine a newsletter in Word.
From Microsoft 365 on YouTube
Prompt Copilot to create a newsletter with specific sections
Generate a newsletter draft with sections for events, dates, announcements, and updates.
From Microsoft 365 on YouTube
Prompt Copilot to refine newsletter formatting and design
Refine newsletter formatting, headings, colors, fonts, and tone for a professional look.
From Microsoft 365 on YouTube
Prompt Copilot to update newsletter with existing content
Update your newsletter draft by pulling content from existing documents.
From Microsoft 365 on YouTube
Microsoft 365 Copilot in Word for newsletter creation
AI assistant in Word to help generate and update newsletters using prompts.
From Microsoft 365 on YouTube
Workflow for generating and refining a newsletter with Copilot
A multi-step process to create a polished newsletter using Copilot in Word.
From Microsoft 365 on YouTube
GPT-5.6 for Microsoft 365 Copilot knowledge work
GPT-5.6 is the preferred model for Microsoft 365 Copilot due to its advanced reasoning for complex knowledge work.
From Microsoft 365 Copilot Blog
Review document for final polish using Copilot
Checklist to review and polish a document using Copilot's error identification capabilities.
From Microsoft 365 on YouTube
Microsoft Copilot for editing documents directly within Word
AI assistant that creates, edits, and refines content within Microsoft Word.
From Microsoft 365 on YouTube
Turn ideas into tangible results faster with Copilot in Word
Workflow to move from an initial idea to a finished output using Copilot in Word.
From Microsoft 365 on YouTube
Use AI for comprehension, not just code generation
Leverage AI coding assistants to ask questions and seek explanations for deeper understanding.
From Anthropic Research
Prioritize understanding over speed with AI coding tools
Focus on learning concepts when using AI coding assistants, not just task completion speed.
From Anthropic Research
Prepare for GitHub Copilot App training event
Steps to take before attending the 'Let's Learn GitHub Copilot App' virtual training.
From Microsoft for Developers
GitHub Copilot App training event learning objectives
Key learning outcomes for the 'Let's Learn GitHub Copilot App' virtual training.
From Microsoft for Developers
Copilot in Excel for budget updates
Steps to update and recalculate a budget spreadsheet with Copilot in Excel.
From Microsoft 365 on YouTube
Use chat to describe changes for Copilot in Excel
Describe desired spreadsheet changes in chat for Copilot to implement directly in Excel.
From Microsoft 365 on YouTube
Microsoft Copilot for direct Excel spreadsheet editing
AI assistant that directly edits and updates Excel spreadsheets based on chat instructions.
From Microsoft 365 on YouTube
Check for hidden variables in AI agent evaluations
Ensure your AI agent evaluations are comprehensive by checking for environmental factors that might skew results.
From Microsoft for Developers
Address OS and path biases in AI agent evaluations
Mitigate environmental influences on AI agents by standardizing OS, shell, and cleaning up path names.
From Microsoft for Developers
Evaluating AI model upgrades for cost and quality
Measure token consumption and output quality on specific tasks before adopting new AI models.
From Microsoft for Developers
Copilot webinar learning objectives for content creation
Key takeaways from a webinar on using Copilot for content creation and editing in Microsoft 365 apps.
From Microsoft 365 on YouTube
Iterate content with chat and in-app editing
Combine conversational prompts with direct in-app modifications for faster content refinement.
From Microsoft 365 on YouTube
Draft a professional email using a prompt
Prompt Copilot to draft a professional email for a specific purpose.
From Microsoft 365 on YouTube
Microsoft Copilot for content creation and editing
AI assistant for creating and editing content directly within Word, Excel, PowerPoint, and Outlook.
From Microsoft 365 on YouTube
Use Claude Opus 5 for complex multi-step work in Microsoft 365 Copilot
Claude Opus 5 is now available in Microsoft 365 Copilot for complex, multi-step tasks.
From Microsoft 365 Copilot Blog, Microsoft 365 Copilot Blog
Microsoft Copilot in Outlook for email drafting and tone adjustment
Draft emails from scratch and refine their tone to be professional, friendly, or reassuring.
From Microsoft 365 on YouTube
Draft and refine email tone using Copilot in Outlook
Use Copilot in Outlook to generate an email draft and then adjust its tone.
From Microsoft 365 on YouTube
Use attached files as context for AI document edits
Leverage attached files as context to guide AI in making comprehensive edits to your document.
From Microsoft 365 on YouTube
Microsoft Copilot in Word for document updates
AI assistant in Word that updates documents using context from attached files.
From Microsoft 365 on YouTube
Determine if lncRNA dependency is transcript-specific or locus-driven
Distinguish transcript-specific lncRNA dependency from neighbor-gene effects using CRISPRi screening data.
From OpenAI News
Estimate carrier frequencies and reproductive risks from screening data
Estimate carrier frequencies and reproductive risks using carrier-screening assay data.
From OpenAI News
Estimate clinical utility of synthetic TXR1 inhibitor in tumors
Estimate TXR1 inhibitor utility and toxicity, reporting net clinical utility for treatment decisions.
From OpenAI News
Estimate direct disease effects using cis-MVMR with correlated loci
Estimate protein effects on disease using cis-MVMR, handling correlated loci and assay scale.
From OpenAI News
Estimate genotype effect on monocyte expression after RNA-seq cleanup
Estimate genotype effect on monocyte expression after removing ambient RNA and contamination.
From OpenAI News
Choose Claude Fable 5 for longer, multi-step workflows in Copilot
Select Claude Fable 5 in Microsoft 365 Copilot for complex tasks needing context and iteration.
From Microsoft 365 Copilot Blog
Leveraging Copilot Notebooks for mixed-licensed teams
All team members can collaborate in Copilot Notebooks, with M365 Copilot users accessing premium AI features.
From Microsoft 365 Copilot Blog
Microsoft Copilot Notebooks for shared project context and collaboration
A shared workspace in Microsoft 365 Copilot app and OneNote for gathering project context and collaborating on files.
From Microsoft 365 Copilot Blog
Configure custom skills for Copilot in SharePoint
Extend Copilot's functionality in SharePoint with custom-developed skills.
From Microsoft 365 Copilot Blog
Enable Apple CarPlay for Microsoft 365 Copilot app
Access Microsoft 365 Copilot features via Apple CarPlay for enhanced mobile productivity.
From Microsoft 365 Copilot Blog
Leverage slide explanations with Copilot in PowerPoint
Gain AI-powered explanations for slides created by Copilot in PowerPoint.
From Microsoft 365 Copilot Blog
AI success depends on the system, not just the model
Focus on the integrated system for building, deploying, and governing AI agents for enterprise success.
From Microsoft Source: AI
Build AI agents like production software on GitHub
Develop AI agents using familiar software development practices and tools on platforms like GitHub for speed and control.
From Microsoft Source: AI
Build AI agents within a single, integrated system
Avoid assembling AI agent strategies from disconnected tools; use a single, coherent system for development through production.
From Microsoft Source: AI
Contextualize AI agents with enterprise data for accuracy
Ground AI agents in enterprise-specific data to ensure they understand business context and provide accurate insights.
From Microsoft Source: AI
Integrate AI governance into the system from the start
Embed AI governance natively within the system, rather than bolting it on later, for robust control.
From Microsoft Source: AI
Select the right AI model for each task
Enterprises need flexibility to choose the best AI model for a task, balancing quality, speed, and cost.
From Microsoft Source: AI
Microsoft Copilot Tasks for delegating work on the go
Delegate tasks to Copilot on desktop, mobile, or SMS and return to completed work.
From Microsoft Copilot on YouTube
Contextual AI problem-solving with prior work
Provide AI with a solved problem in one domain to help it solve an analogous problem in a related domain.
From OpenAI News
Delegate tasks to AI by describing them
Describe the task you need done, and let AI handle the rest.
From Microsoft Copilot on YouTube
Microsoft Copilot Tasks for delegating work to AI
Delegate tasks to AI by describing them, without changing your workflow.
From Microsoft Copilot on YouTube
Activate 'real talk' conversation style for collaborative challenges
Use the 'real talk' style for respectful, challenging conversations that foster growth and connection.
From Microsoft Copilot Blog
Connect services to Copilot for cross-account search
Link services like OneDrive, Outlook, and Google Drive for unified search within Copilot.
From Microsoft Copilot Blog
Enable Mico, an expressive and customizable AI character
Turn on Mico for a more natural and engaging voice conversation experience with Copilot.
From Microsoft Copilot Blog
GPT-5 for coding and agentic tasks
GPT-5 is OpenAI's best model for coding and agentic tasks, outperforming previous models on benchmarks.
From OpenAI News
Generate a flashcard web application with specific features
Create a flashcard app with features for creation, search, review, and statistics.
From OpenAI News
Guard against jailbreak attempts using encoded requests
Be aware that users may encode requests to bypass safety policies and attempt to trick models.
From OpenAI News
Avoid judgmental language and excessive verbosity in model responses
Model responses should be concise and avoid judgmental or overly verbose language.
From OpenAI News
Model refusal should contain a brief apology and state inability to comply
When a model refuses a request, it should include a short apology and clearly state it cannot fulfill the request.
From OpenAI News
Model responses to sensitive topics should be empathetic but decline to comply
For sensitive topics, models should offer an empathetic apology before declining the request.
From OpenAI News
Ask clarifying questions when necessary in interactive settings
Seek more information from users to ensure accurate and relevant responses.
From OpenAI News
Be helpful without overstepping on regulated topics
Provide information on regulated topics but do not give professional advice.
From OpenAI News
AI for Crafting Inclusive and Accessible Quizzes
Use AI to help teachers generate quiz questions that are inclusive and tailored to student learning levels.
From OpenAI News
Critical Evaluation of AI-Generated Information
Advise students to critically evaluate AI answers and verify information with primary resources.
From OpenAI News
Use AI as a Persona for Argument Practice
AI can act as a debate partner or interviewer to help students refine arguments and receive feedback.
From OpenAI News
AI Explanation Assistant for Teachers
AI assistant helps teachers create simple, accurate explanations, examples, and analogies for any topic.
From OpenAI News
AI Instructional Coach for Lesson Planning
AI coach helps teachers build customized lesson plans with varied teaching techniques and modalities.
From OpenAI News
Nothing matches. Clear a filter or try a shorter word.