# Heighliner

> Heighliner runs the AI agents you already use as fleets, with a dollar budget that hard-stops spend, approvals for risky actions, a sandbox per agent and a tamper-evident audit trail.

- Status: Early access. Not generally available yet; join the waitlist at https://heighliner.run/waitlist.
- What it is: infrastructure for running AI agent fleets, not the agents themselves.
- Bring your own agent: Claude Code, Codex, Hermes, OpenCode, Custom container, Your own endpoint.
- Every fleet gets: budgets, schedules and triggers, sandboxes, approvals, audit.
- Bring your own model key (any provider or OpenRouter). Tokens are tracked for budgets and never marked up.
- Agent-first API: plan, quote, apply. MCP connector for ChatGPT and Claude, the hl CLI, and TypeScript, Python and Go SDKs.
- Deployment: Managed cloud (From $0); BYOC (From about $36k a year); Self-hosted (Annual license, contact us). All from one codebase.
- Managed plans: Free $0, Pro $20/month, Team $200/month, Enterprise Custom. Early access pricing; subject to change.
- Docs: https://docs.heighliner.run
- Feed: https://heighliner.run/rss.xml (RSS, blog posts newest first)

## Pages
- [Home](https://heighliner.run/): Heighliner runs the AI agents you already use as fleets, with a dollar budget that hard-stops spend, approvals for risky actions, a sandbox per agent and a tamper-evident audit trail.
- [Product](https://heighliner.run/product): What every Heighliner fleet gets: a dollar budget with a hard stop, schedules and triggers, a sandbox per agent, approvals for risky actions and an audit trail.
- [How it works](https://heighliner.run/how-it-works): How Heighliner runs an agent fleet: describe it in a manifest, plan and quote every change, apply it, and govern risky actions with approvals.
- [Pricing](https://heighliner.run/pricing): Heighliner early access pricing: Free, Pro $20 and Team $200 with usage credit, per-second usage, model tokens on your own key, plus BYOC and self-hosted.
- [Deployment](https://heighliner.run/deployment): Run Heighliner three ways from one codebase: our managed cloud, your own AWS, GCP or Azure account (BYOC), or fully self-hosted and air-gapped.
- [Security](https://heighliner.run/security): How Heighliner contains AI agents: a sandbox per agent, keys outside the sandbox, egress through a gateway, approvals with escrow and a hash-chained audit trail.
- [Developers](https://heighliner.run/developers): The hl CLI, TypeScript, Python and Go SDKs, the MCP connector for ChatGPT and Claude, and an agent-first API built on plan, quote and apply.
- [About](https://heighliner.run/about): Why Heighliner exists: infrastructure for running AI agent fleets, built to sell trust rather than speed, for developers first.
- [Early access](https://heighliner.run/waitlist): Join the Heighliner early access waitlist to run your AI agents with budgets, approvals, sandboxes and an audit trail.
- [Changelog](https://heighliner.run/changelog): What has changed at Heighliner, newest first.
- [Blog](https://heighliner.run/blog): Notes on budgets, approvals, sandboxes and audit trails for people who run AI agents on real work, from the team building Heighliner.
- [Use cases](https://heighliner.run/use-cases): Agents people run on a schedule today, and what a budget, approvals and an audit trail change for each: SEO, research, PR review and reports.
- [Compare](https://heighliner.run/compare): Honest comparisons between Heighliner and other ways to run AI agents, including a VPS with cron and hosted agent sandboxes.
- [Glossary](https://heighliner.run/glossary): Plain definitions of the terms used across Heighliner: fleets, budget leases, effect classes, escrow, sandboxes and BYOC.

## Blog
- [Why agents need a budget, not a dashboard](https://heighliner.run/blog/why-agents-need-a-budget-not-a-dashboard): A dashboard tells you what an agent spent after the fact. A budget stops it before it overspends. How dollar caps and leases make agents safe to leave running.
- [Approvals without rubber-stamping](https://heighliner.run/blog/approvals-without-rubber-stamping): Asking a person to approve every agent action turns oversight into clicking yes. Effect classes, escrow and saved rules keep approvals rare and meaningful.

## Use cases
- [Competitor research agents that report back and shut down](https://heighliner.run/use-cases/competitor-research): Send several agents to research competitors in parallel, each in its own sandbox, under one dollar budget, and get a single brief with a receipt for how it was made.
- [Nightly reports that are drafted automatically and sent deliberately](https://heighliner.run/use-cases/nightly-reports): Run an agent every night to pull metrics and draft a report. Sending email to customers is irreversible, so it waits in escrow for a person to approve.
- [A pull request reviewer that comments freely and merges never](https://heighliner.run/use-cases/pr-reviewer): Run Codex, Claude Code or your own agent as a pull request reviewer triggered by webhook, with scoped repository access, a budget and an audit trail of every comment.
- [An SEO agent that runs every week and cannot overspend](https://heighliner.run/use-cases/seo-agent): Run Claude Code, Codex or Hermes as a scheduled SEO agent that opens pull requests for fixes, with a dollar cap and approval before anything merges.

## Compare
- [Heighliner vs hosted agent sandboxes](https://heighliner.run/compare/hosted-agent-sandboxes): Hosted sandbox services give an agent an isolated machine through an API. Heighliner runs whole fleets with budgets, approvals and audit around them. Which do you need?
- [Heighliner vs a VPS, a cron job and a .env file](https://heighliner.run/compare/vps-and-cron): An honest comparison of running AI agents on your own VPS with cron and running them on Heighliner: cost, control, budgets, approvals and when the VPS is the better choice.

## Glossary
- [Budget lease](https://heighliner.run/glossary/budget-lease): A budget lease is a small slice of a fleet's dollar budget, $1 by default, that the gateway spends from locally, so caps are enforced on every call without slowing it.
- [BYOC](https://heighliner.run/glossary/byoc): BYOC, bring your own cloud, means Heighliner's data plane runs in your AWS, GCP or Azure account, so agents, data and traces stay there while we run the control plane.
- [Effect class](https://heighliner.run/glossary/effect-class): An effect class labels every agent action by how hard it is to undo: pure, reversible, compensatable or irreversible. Policy and approvals follow from it.
- [Escrow](https://heighliner.run/glossary/escrow): Effect escrow stages an agent's irreversible actions, like sending email, in an outbox and releases them only after approval and when the run commits.
- [Fleet](https://heighliner.run/glossary/fleet): A fleet is one or more AI agents running one harness under one manifest, one budget and one policy. A single scheduled agent is a fleet of one.
- [Sandbox](https://heighliner.run/glossary/sandbox): On Heighliner every agent runs in its own sandbox, with keys attached outside it and network egress only through the gateway, so one agent cannot reach another's work.

## FAQ
### Is Heighliner an AI agent?
No. Heighliner is the infrastructure agents run on. You bring the agent, such as Claude Code, Codex, Hermes, OpenCode, a custom container or your own endpoint, and Heighliner adds budgets, schedules, sandboxes, approvals and an audit trail.

### Who is it for?
Developers and indie hackers who run agents today on a VPS with a cron job and a .env file, and agencies running agents for clients. Enterprises can run it in their own cloud or fully self-hosted.

### Which models can I use?
Any provider, or OpenRouter, with your own key. Heighliner tracks tokens so budgets work, and never marks them up.

### Can I keep my data in my own cloud?
Yes. With BYOC the data plane runs in your AWS, GCP or Azure account and your data stays there. Fully self-hosted and air-gapped installs are also available.

### Is it available now?
Heighliner is in early access. Join the waitlist and we will reach out as places open.