How GitHub expanded secret validity checks with Copilot coding agent
GitHub's Secret Protection team had engineers do the research, then gave Copilot coding agent the repetitive work of adding validators for more leaked-token types. Coverage grew quickly.
The problem
Secret Protection checks whether a leaked credential is still valid by testing it against a harmless API endpoint. Only 32 token types were covered, and adding less common types through the existing framework was slow.
How they did it
- Engineers researched which API endpoints could validate each new token type.
- A GitHub workflow turned that research into detailed issues that served as prompts for the agent.
- Copilot coding agent opened pull requests, and engineers reviewed, tested and asked for changes through PR comments.
- New validators first ran in 'darkship' mode to observe results safely.
- Copilot then made the configuration changes to fully release them.
Results as reported by GitHub
- As reported by GitHub: the team went from validating '32 partner token types' to onboarding 'almost 90 new types in just a few weeks'.
These figures are as reported by GitHub in GitHub blog (engineering) ↗ (Aug 28, 2025). GitHub sells the product used here. AgentGid did not measure or independently verify them.
This suits a team with engineers who can research each item and write detailed issues, plus the review capacity to test every PR. The "almost 90 types in a few weeks" figure is GitHub's own report, and the cloud agent draws on monthly AI Credits (Pro includes 1,500 at $10/mo), so heavy batches of issues may use them up. Zed (Free + $10/mo) is an open-source option if you want to try a similar workflow with your own API keys.
The agent used here
Similar use cases
League halved product cycle times with Claude Code and overnight agent teams
League, which builds member health apps for health plans and providers, rolled Claude out company-wide and moved its engineers to Claude Code. Small pods now ship production code …
Cycle times from idea through pull request cut in half in the four months after the Accelatron, as reported by Anthropic.
How three Java Champions use GitHub Copilot for migration, testing and app building
Three experienced Java developers describe how they use GitHub Copilot in IDEs and agent workflows for feature work, a slide-deck migration, UI test reproduction and small apps. T…
As reported by the GitHub Blog (developer quotes): Vlad Mihalcea built an Optimizer event-history web app in two days, a task he expected to take wee…
Generate unit and integration tests with GitHub Copilot
GitHub's tutorial uses Copilot Chat to write a unit test suite for a class, then integration tests that mock an external notification service, and then asks what coverage is still…
A unit test suite that covers edge cases, exceptions and data validation. (as reported by GitHub)
Guides
Best AI Code Review Tools in 2026: CodeRabbit, Greptile, Graphite, Copilot and Open Source
AI code review agents for pull requests compared: what they check, how they are priced, open-source options and how to roll one out without drowning your team in comments.
AI Coding Agent Pricing Explained: Claude Code, Cursor, Codex and Copilot (2026)
What Claude Code, Cursor, OpenAI Codex, GitHub Copilot and other coding agents really cost: plan prices, usage limits, subscription vs API billing, and how to estimate your monthly bill.
What Is MCP (Model Context Protocol)? A Plain Guide and the Most Popular MCP Servers
What the Model Context Protocol is, how MCP servers and clients work, which AI agents support it, how to install a server safely, and the most-starred MCP servers by category.
Best AI Agents in 2026: Live Rankings by Category
The leading AI agents for coding, design, research, data, marketing, sales, support, HR, finance, legal, DevOps and security, ranked by public usage data that refreshes daily, with prices and free options.