Distinguish a code suggestion from an agent that can edit files and run commands.
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.
The worked example is illustrative and has not been tested in a signed-in product account.
What it does and how it connects
Distinguish a code suggestion from an agent that can edit files and run commands.
Completions suggest code near the cursor; chat explains or proposes changes; agent mode can choose files and run commands. Editor context is not necessarily the entire repository, and product support varies by IDE. GitHub Copilot has its own plans and AI credit accounting, separate from Microsoft 365 Copilot.
Walkthrough
Use the product named in the access panel. The steps below combine documented controls with the editorial example.
- Install / enable Copilot through the official instructions for your supported IDE.
- Open a sample file and ask a focused explanation question.
- Review a suggestion before accepting it; compare it with the project’s conventions.
- Use agent mode only when the task needs a multi-step change.
Worked example
Ask for an explanation of a small input validator, then request a suggestion for handling an empty string. Inspect it manually and test known input cases before accepting the suggestion into the repository.
Explain this validator using the actual code. Identify how it handles empty input and unexpected types. Suggest one scoped improvement, but do not edit yet. Give example inputs and expected outcomes I can independently test.
Inspect and verify the result
Read the relevant code and run the input cases. Check whether suggested functions or packages actually exist. Compare error behavior and types with the surrounding code.
Limits and troubleshooting
Context, model access, and support differ by client. A convincing completion can use nonexistent APIs. If the explanation refers to missing files, add the intended context explicitly rather than assuming whole-repo knowledge.
Data, permissions, and actions
Code context and prompts are processed under the selected account’s terms and policy. Organizational controls and source permissions still matter. Suggestions do not authorize shell execution, repo writes, or deployment.
Related reading
IDE agent mode and tool approvals
Use an interactive agent for a bounded multi-file change and review each consequential command.
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.
Evaluations: does the system do the job?
Create representative tests for quality, tool behavior, refusal, escalation, and the effects of a change.
Sources and verification
Documentation checked 2026-10-04. Examples are illustrative unless described otherwise.
About GitHub Copilot · GitHub
Plans for GitHub Copilot · GitHub