From one-off prompts to workflows: How to use custom agents in GitHub Copilot CLI

2026-06-09 · Source: The GitHub Blog · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy · Depth: Intermediate, long

Summary

The GitHub Copilot CLI now supports custom agents, introduced on June 9, 2026, enabling developers to transform one-off terminal prompts into repeatable, reviewable workflows. These agents are defined via Markdown files, called agent profiles, which reside in a repository's ".github/agents" directory. This allows teams to encode their specific stack, tools, and operational standards, ensuring consistent behavior across development tasks. Agent profiles specify the agent's role, accessible tools, and guardrails, supporting version control and sharing. The article illustrates practical applications, including agents for security audits, Infrastructure-as-Code compliance, release documentation, and incident response. While off-the-shelf agents offer quick starts, custom agents provide tailored integration with unique team conventions and internal tooling, enhancing workflow continuity from CLI to IDE and GitHub.

Key takeaway

For DevOps Engineers aiming to standardize and automate complex terminal-based workflows, custom agents in GitHub Copilot CLI offer a powerful solution. You should define agent profiles in Markdown within your repositories to encode team-specific tools, standards, and guardrails. This approach ensures consistent execution of tasks like security audits or IaC compliance, reducing manual errors and improving reviewability across your development lifecycle. Start by converting a frequently repeated task into a custom agent.

Key insights

Custom agents in GitHub Copilot CLI standardize development workflows by encoding team-specific context into reusable, version-controlled automation.

Principles

• Encode team context into reusable workflows.
• Define agent behavior with Markdown profiles.
• Version control agents like other code.

Method

To create a custom agent: invoke Copilot CLI with "/agent", then create a Markdown agent profile in ".github/agents/" defining its role, tools, and guardrails.

In practice

• Automate security audits with team-specific checks.
• Enforce IaC compliance against organizational policies.
• Draft release notes and update CHANGELOG.md.

Topics

• GitHub Copilot CLI
• Custom Agents
• Workflow Automation
• DevOps Tools
• Infrastructure as Code
• Security Audits
• Incident Response

Code references

• features/copilot
• github/awesome-copilot

Best for: Software Engineer, AI Engineer, DevOps Engineer

Related on AIssential

• AI Agents Need Their Own Desk, and Git Worktrees Give Them One — Git worktrees enable parallel development by providing isolated working directories for different branches, crucial for AI Agent workflows.
• Google Cloud Introduces Agents CLI to Streamline AI Agent Development Lifecycle — Agents CLI unifies AI agent development, simplifying prototyping, evaluation, and deployment on Google Cloud.
• Mitchell Hashimoto’s new way of writing code — AI agents are transforming software development workflows and challenging traditional open source and version control paradigms.
• Build an agent that writes its own tools — Agents can safely author their own tools from raw API specs within a secure sandbox, eliminating manual tool wrapper maintenance.
• Subagents in Gemini CLI Enable Task Delegation and Parallel Agent Workflows — Gemini CLI subagents enable task delegation and parallel execution for complex AI workflows, reducing context overload.
• Rethinking Git for the Age of Coding Agents with GitHub Cofounder Scott Chacon — Git's original design and backward compatibility limit its utility for modern AI agent workflows, necessitating new version control approaches.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The GitHub Blog.