RelayRunner docs: Keep up with coding agents Page: https://relayrunner.dev/docs/guides/agent-surge Markdown: https://relayrunner.dev/docs/guides/agent-surge.md All pages: https://relayrunner.dev/docs/llms.txt Other pages (Markdown): - Route GitHub Actions jobs to any runner: https://relayrunner.dev/docs/get-started/overview.md - How it works with GitHub Actions: https://relayrunner.dev/docs/get-started/github-actions.md - Quickstart: https://relayrunner.dev/docs/get-started/quickstart.md - Install the CLI: https://relayrunner.dev/docs/get-started/install.md - RelayRunner for Mac: https://relayrunner.dev/docs/get-started/mac-app.md - Set up with an agent: https://relayrunner.dev/docs/get-started/agents.md - Move your workflows over: https://relayrunner.dev/docs/get-started/move-over.md - Principles: https://relayrunner.dev/docs/concepts/principles.md - Routing: https://relayrunner.dev/docs/concepts/routing.md - Runner order: https://relayrunner.dev/docs/concepts/runner-order.md - Optimized checks: https://relayrunner.dev/docs/concepts/optimized-checks.md - Spend controls: https://relayrunner.dev/docs/guides/spend-controls.md - Providers: https://relayrunner.dev/docs/runners/providers.md - Your machines: https://relayrunner.dev/docs/runners/self-hosted.md - GitHub-hosted: https://relayrunner.dev/docs/runners/github.md - Hosted providers: https://relayrunner.dev/docs/runners/hosted.md - Runner labels: https://relayrunner.dev/docs/reference/labels.md - Add a provider: https://relayrunner.dev/docs/contribute/add-provider.md - Run CI on your laptop while you work: https://relayrunner.dev/docs/guides/laptop-ci.md - macOS and iOS builds on your Mac: https://relayrunner.dev/docs/guides/mac-builds.md - Share your team's machines: https://relayrunner.dev/docs/guides/team-machines.md - Keep up with coding agents: https://relayrunner.dev/docs/guides/agent-surge.md (this page) - Activity and history: https://relayrunner.dev/docs/dashboard/activity.md - Signals: https://relayrunner.dev/docs/dashboard/signals.md - Performance: https://relayrunner.dev/docs/dashboard/performance.md - Usage and savings: https://relayrunner.dev/docs/dashboard/usage.md - Telemetry: https://relayrunner.dev/docs/dashboard/telemetry.md - AI: https://relayrunner.dev/docs/dashboard/ai.md - CLI: https://relayrunner.dev/docs/reference/cli.md - relay.yml inputs: https://relayrunner.dev/docs/reference/relay-inputs.md - Repo config: https://relayrunner.dev/docs/reference/repo-config.md - MCP server: https://relayrunner.dev/docs/reference/mcp.md - RelayRunner API: https://relayrunner.dev/docs/reference/api.md - Workspaces and members: https://relayrunner.dev/docs/dashboard/workspaces.md - GitHub App: https://relayrunner.dev/docs/reference/github-app.md - Security: https://relayrunner.dev/docs/reference/security.md - Data and privacy: https://relayrunner.dev/docs/reference/privacy.md - Remove RelayRunner: https://relayrunner.dev/docs/dashboard/remove.md - Troubleshooting: https://relayrunner.dev/docs/reference/troubleshooting.md - Errors: https://relayrunner.dev/docs/reference/errors.md - FAQ: https://relayrunner.dev/docs/reference/faq.md - Feedback: https://relayrunner.dev/docs/contribute/feedback.md - Changelog: https://relayrunner.dev/docs/changelog.md - Agents can change your routing: https://relayrunner.dev/docs/changelog/agent-changes.md - Pick runner sizes on the dashboard: https://relayrunner.dev/docs/changelog/auto-runner-sizes.md - CLI changelog: https://relayrunner.dev/docs/changelog/cli.md - Send job logs to your tools or your own bucket: https://relayrunner.dev/docs/changelog/job-log-export.md - Mac app changelog: https://relayrunner.dev/docs/changelog/mac.md - RelayRunner for Mac: https://relayrunner.dev/docs/changelog/mac-app.md - CI traces in your observability tool: https://relayrunner.dev/docs/changelog/opentelemetry-traces.md - Platform changelog: https://relayrunner.dev/docs/changelog/platform.md - Retries, queueing and provider usage: https://relayrunner.dev/docs/changelog/retries-queueing-and-provider-usage.md - Route a repo from the dashboard: https://relayrunner.dev/docs/changelog/route-a-repo.md - Clearer run times and job sizes for members: https://relayrunner.dev/docs/changelog/run-times.md - Usual times, job history and the Mac queue: https://relayrunner.dev/docs/changelog/usual-times-and-mac-queue.md --- # Keep up with coding agents > When agents push all day, run CI on your own machines first and burst to hosted runners when they're full. Coding agents open pull requests around the clock, and each push runs CI. A month's included minutes can go in a few days, a single self-hosted machine turns into a queue, and paying for fast hosted runners for every push adds up quickly. This guide uses every machine you have first, and sends only the overflow to a hosted provider. ## 1. Connect what you have Run `relayrunner connect` on every spare computer: the Mac mini, the old laptop, the home server. Each one runs a job or two at once (more with `--concurrency`), so more machines means more jobs at once. Share them with **anyone** in the workspace so jobs from agents and teammates can use them: ```sh relayrunner share anyone ``` Jobs started by a bot only run on machines shared with anyone. ## 2. Burst to hosted runners Set the runner order so the overflow goes to a provider instead of waiting: 1. **Self-hosted** 2. **Blacksmith** (optional, if you want faster hosted runners) 3. **GitHub-hosted** When every machine is busy, the next job goes to the next step straight away, instead of queueing behind the others. ## 3. Cap the hosted spend Put a minutes limit on GitHub so overflow stops billing at a point you choose. See [Spend controls](/docs/guides/spend-controls). With no unlimited step after it, a job that finds every machine busy fails fast with the reason, and the agent can retry. ## 4. Take the cheap jobs off runners PR title checks and lint don't need a machine at all. Make them [optimized checks](/docs/concepts/optimized-checks) so they never wait for a runner or bill a minute. ## Watch it The **Performance** page shows how long each job takes on each runner and how long it waited for one. If waits grow, connect another machine or move a step up.