Jetty documentation
Jetty is where you find, share, and run reliable AI workflows. You write a runbook — one markdown file that gives an agent its job, its bar for “done”, and its checks — and Jetty runs it in a managed sandbox, records every run in full, and helps you improve it until it works every time.
These docs are written for builders: developers who live in Claude Code, Codex, or Cursor and want a runbook running in minutes instead of a framework to stand up. Pick a starting point.
Get started
From zero to a runbook you trust, in one sitting.
What Jetty runs, and three ways in: start free, install locally, or call it from your app.
Browse the Jetty Runbook Directory and try it yourself on your own account.
Install Jetty locally: the plugin, /jetty-setup, and your first run.
Every on-ramp: the MCP server, the Client SDK, and the API.
Product overview
What Jetty does, in two motions: run the workload, then read the run.
The two halves of Jetty: run AI workloads in managed sandboxes, and investigate how well the agent performed.
How a runbook becomes a running workflow: the author → run → investigate → improve lifecycle, and what Jetty assembles around every run.
Read a run end to end: the trajectory, the workload report, and the loop that turns findings into fixes.
AI workloads
The job you hand to Jetty, and the pieces that carry it out.
What counts as an AI workload, and the two pieces every one is built from: a runbook and a runtime.
The hero artifact: one markdown file with the job, the bar for “done”, and the checks.
The agents Jetty runs inside its own sandboxes (claude-code, codex, gemini-cli, antigravity, opencode, hermes, pi, goose), which providers back each, and how to pick one.
Provider keys and environment variables, and how sandboxed runs read them.
Every run is captured end to end. Evals turn runs into a hill-climbing loop with /optimize-runbook.
One API, two modes (passthrough proxy and runbook sandbox), durable execution, object storage.
Jetty and Friends
Already have an agent in another framework? Keep it. Jetty becomes the independent evaluation, observability, and optimization layer around it, or you run one of the four built-in runtimes inside Jetty's own sandboxes.
Keep your agent in any framework; use Jetty as the independent grader, store, and A/B harness. Flue and eve worked examples.
Keep your agent in its own framework and let Jetty grade every run: the pattern, with Eve and Flue worked examples.
Mount the @jetty/eve extension into a Vercel eve agent: live grading, a grade-steered experiment bandit, and an eval reporter. Catch a regression before it ships.
The complete flue-jetty walkthrough: grade every Flue agent run with an independent rubric and catch a regression before it ships.
@jetty/sdk for TypeScript: JettyClient, runAndWait, and the run → check → fix → rerun loop.
npx -y jetty-mcp-server — 16+ tools across Cursor, VS Code, Windsurf, Zed, Codex.
Go deeper
The task-shaped guides, when you want them.
The canonical structure: frontmatter, objective, output manifest, evals, and the /create-runbook wizard.
Add evals to a runbook, read the runs, and hill-climb with /optimize-runbook.
Run a runbook on a cron cadence so evals stay fresh as models and providers drift.
The chat-completions endpoint (two modes), routines, webhooks, and GitHub-PR APIs.
The runbook API surface for agents and programmatic callers: tasks, runs, init_params, trajectories, files, MCP config, sweeps, snapshots, webhooks, routines.
The catalog of activities: AI models, control flow, data processing, evaluation.
Ready to build?
Get started → — token to first runbook in under ten minutes.
Prefer to talk it through? Book a 20-minute walkthrough → Bring a task from your own work and we'll turn it into a runbook together.