An editor and repository assistance family spanning suggestions, interactive agents, cloud work, and code review.
Coding & development · Documented · Documentation checked 2026-10-04
Editor assistance and agents, GitHub cloud agent, and code review have different execution contexts.
Plan, supported IDE, repository permission, AI credits, and organization policy determine access.
What the platform is for
GitHub Copilot can suggest code, explain a file, run an interactive agent task, or create a cloud-agent handoff. These are different degrees of tool use. Begin with an explanation and a checked suggestion, then progress toward agentic edits only when the task requires multiple steps.
Know the product boundary
Completions, IDE chat / agent mode, GitHub cloud agent, and code review run in different contexts. IDE support, repo access, organization policy, plan, and AI credits affect availability. Microsoft 365 Copilot is a separate product. Cloud access does not mean access to a developer’s current laptop state.
How the components connect
Instructions and custom agents describe task conventions. Model choice and editor context shape the response. MCP tools add external authority. Interactive agents change the workspace; cloud agents produce reviewable branches or pull requests. Code review raises findings, which still need reproduction and human judgment.
A beginner starting point
Start with a known function and test its explanation. Accept a suggestion only after checking it. Use a scoped IDE agent change next, then repository instructions and an approved tool. Approach cloud agent sessions as a pull-request handoff, with human review and normal branch protection.
Limits and data choices
Suggestions can invent APIs, reviews can miss defects, and cloud checks can lack local dependencies. Instruction support differs by feature and surface. This edition avoids fixed price / quota promises and does not exhaust every CLI, partner agent, API automation, or enterprise policy.
The technical path
Use official GitHub Docs for plan tables, editor-specific setup, cloud-agent APIs, custom agents, code-review configuration, and administration. Existing CI, branch rules, deployment credentials, and source permissions remain the project’s enforcement layers.
Feature map
Completions, chat, and editor context
Distinguish a code suggestion from an agent that can edit files and run commands.
IDE agent mode and tool approvals
Use an interactive agent for a bounded multi-file change and review each consequential command.
Instructions, custom agents, and MCP
Combine repository guidance with explicitly authorized tools, without treating instructions as security controls.
Cloud agent and pull request handoff
Turn a well-defined repository issue into a reviewable branch and pull request.
Code review, plans, and AI credits
Check suggested findings against the code and understand the account budget used by agentic features.
Complete workflows
An editor fix with a checked handoff
Move from a checked suggestion to a scoped IDE agent change.
From a clear issue to a reviewed pull request
Create a cloud-agent change and a human-reviewed pull-request handoff.
Coverage and availability
Models, modes & conversation
Covered: Suggestions, chat, Agent mode, and available model choices.
Search, research & sources
Covered: Repository / connected context; consumer Copilot Researcher is separate.
Files, analysis & code
Covered: Code edits, commands, checks, and review.
Images, video, audio & voice
Covered: Screenshot context may be supported; not a consumer video / audio generator guide.
Projects, instructions & memory
Covered: Repository instructions and custom agents; support varies by surface.
Editing surfaces & deliverables
Covered: Editor changes and pull-request handoff.
Agents, skills, plugins & MCP
Covered: IDE agents, cloud agent, custom guidance, and MCP.
Orchestration & scheduling
Not established: Cloud automation entrypoints documented; production scheduling design deferred.
APIs, SDKs, deployment & evaluation
Separate product: Review and cloud API paths; project CI / deployment remains separate.
Sharing, privacy & governance
Covered: Repository permissions, organization enablement, and merge control.
Plans, devices & accessibility
Not established: AI credit / plan relationship noted; full quotas, regions, and accessibility not established.
Sources and verification
Documentation checked 2026-10-04. Examples are illustrative unless described otherwise.
About GitHub Copilot · GitHub
Plans for GitHub Copilot · GitHub
Use agent mode in your IDE · GitHub
Repository custom instructions · GitHub
Enhance agent mode with MCP · GitHub
Copilot cloud agent · GitHub
Copilot code review · GitHub