Custom agents bring team-specific workflows into GitHub Copilot

Engineering teams operate on more than code—they operate on conventions. Terraform module structuring, trusted dashboards, database migration rules, security standards: these patterns live across observability, CI/CD, and third-party tools. GitHub Copilot now supports custom agents that encode these workflows directly into the assistant, extending its reach across the full software development lifecycle.

Custom agents are Markdown-defined domain experts that plug into the GitHub Copilot coding agent. They function as reusable teammates: a PagerDuty incident responder, a JFrog security analyst, or a MongoDB performance specialist. Because they're plain Markdown files, there is no separate infrastructure to stand up or maintain.

You can run these agents across every Copilot surface: in the terminal through Copilot CLI, in VS Code via Copilot Chat, and on github.com in the Copilot panel.

Defining a custom agent

An agent definition is a Markdown file. A typical one looks like this:

---
name: readme-specialist
description: Expert at creating and maintaining high-quality README documentation
---

You are a documentation specialist focused on README files. Your expertise includes:

- Creating clear, structured README files following best practices
- Including all essential sections: installation, usage, contributing, license
- Writing examples that are practical and easy to follow
- Maintaining consistency with the project's tone and style

Only work on README.md or documentation files—do not modify code files.

Add the file to your repository's agent directory and the agent appears immediately:

.github/agents/readme-specialist.md
  • GitHub Copilot CLI
  • github.com, in the control plane
  • VS Code in Copilot Chat

Agents can live at two scopes:

  • Repository level: .github/agents/CUSTOM-AGENT-NAME.md for project-specific workflows
  • Organization/Enterprise level: /agents/CUSTOM-AGENT-NAME.md in a .github or .github-private repository, available across all repositories in the organization

Solving real workflows with partner agents

GitHub has worked with partners across the ecosystem to ship pre-built agents targeting common engineering scenarios.

Incident response and remediation

The PagerDuty Incident Responder agent helps triage production incidents directly within your workflow:

copilot --agent=pagerduty-incident-responder \
  --prompt "Summarize active incidents and propose the next investigation steps."
  • Pull context from PagerDuty alerts
  • Generate a clear overview of incident state
  • Recommend investigation paths
  • Draft incident updates for your team

For dependency security, the JFrog Security Agent identifies and remediates vulnerable packages:

copilot --agent=jfrog-security \
  --prompt "Scan for vulnerable dependencies and provide safe upgrade paths."
  • Identify vulnerable packages
  • Provide recommended upgrade versions
  • Patch dependency files directly
  • Generate a security-aware pull request summary

Database and experiment management

The Neon agent handles database workflow modernization:

copilot --agent=neon-migration-specialist \
  --prompt "Review this schema migration for safety and best practices."
  • Validate schema changes
  • Avoid unsafe migrations
  • Tune analytical workflows
  • Optimize transformations and queries

The Amplitude Experiment Implementation agent speeds up product experimentation:

copilot --agent=amplitude-experiment-implementation \
  --prompt "Integrate an A/B test for this feature and generate tracking events."
  • Generate experiment scaffolding
  • Insert clean, consistent event tracking
  • Map variations to product logic
  • Ensure data flows correctly into Amplitude

Why custom agents matter

By codifying your team's patterns and integrations into a reusable agent, Copilot moves from understanding only the code in front of it to understanding how your team works. Custom agents help with:

  • Consistency: Terraform conventions, database rules, and security standards stay uniform across contributions
  • Reduced repetition: Expectations are defined once and reused everywhere
  • Automatic expertise sharing: Best practices propagate even when subject matter experts are unavailable
  • Direct tool integration: Model Context Protocol (MCP) servers feed data from DevOps, security, and observability systems into the agent

The partner agent catalog

Available agents span observability, security, databases, infrastructure, and more.

Observability and monitoring

  • Dynatrace Observability and Security Expert: Configure and optimize Dynatrace monitoring
  • Elasticsearch Remediation Agent: Handle Elasticsearch configuration, query optimization, and observability setup

Security and compliance

  • JFrog Security Agent: Identify and remediate dependency vulnerabilities
  • StackHawk Security Onboarding: Set up dynamic application security testing

Database and data management

  • MongoDB Performance Advisor: Analyze and optimize query performance
  • Neon Migration Specialist: migrate databases to Neon's serverless Postgres
  • Neon Performance Analyzer: Find bottlenecks and optimization opportunities
  • Neo4j Docker Client Generator: Generate Docker-based client code for Neo4j graph databases

DevOps and infrastructure

  • Terraform Infrastructure Agent: Write, review, and optimize infrastructure as code
  • Arm Migration Agent: Migrate applications to Arm-based architectures
  • Octopus Release Notes Expert: Generate release notes from deployment data
  • DiffBlue Java Unit Test Custom Agent: Generate Java unit tests using AI-powered test generation

Incident response and project management

  • PagerDuty Incident Responder: Triage and respond to production incidents
  • Monday Bug Context Fixer: Pull context from monday.com to resolve bugs

Feature management and experimentation

  • LaunchDarkly Flag Cleanup: Identify and remove obsolete feature flags
  • Amplitude Experiment Implementation: Implement A/B tests and experiments

API integration and automation

  • Apify Integration Expert: Integrate web scraping and automation workflows
  • Lingo.dev Internationalization Implementation Custom Agent: Detect, extract, and implement internationalization patterns
  • Factory.ai Code Spec Custom Agent: Automate workflows using Droid CLI for CI/CD pipelines

Run any partner agent with:

copilot --agent=<agent-name> --prompt "<task>"

Availability and next steps

Custom agents are available now for all GitHub Copilot users. Resources for getting started include:

copilot --agent=terraform-agent --prompt "Review my IaC for issues"

Custom agents shift Copilot from "help write this code" to "help build software the way our team builds software"—with your conventions, your tooling, and your standards encoded directly into every session.