> ## Documentation Index > Fetch the complete documentation index at: https://opensre.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Welcome to OpenSRE > Install OpenSRE, run your first CI/CD demo, and scale from a single laptop to a managed gateway for your organization.
Quickstart guide → Create an account →
## Choose a path Start with a CI/CD outcome, an incident question, or a rollout for your organization. Pick the one that matches your goal. A 30-day reliability report for a repository, fixes for failing checks, and an agent that keeps pull requests green. **Run the CI/CD demo →** Paste an alert or describe what is wrong. OpenSRE reads logs, metrics and deploys from your connected tools and answers in plain English. **Run the quickstart →** One account for the organization and a managed gateway that keeps running repairs after the laptop closes. Slack and other integrations are configured on each engineer's machine. **Review the enterprise setup →** ## Start with a surface The same `opensre` binary runs everywhere. Pick a surface, install it, and start your first session. Install with one command, sign in, and pick a demo or type a request. ```bash theme={null} curl -fsSL https://install.opensre.com | bash -s -- -dc opensre ``` Windows uses PowerShell; see [Install](/docs/install). The first prompt offers the CI/CD analysis of a repository on your machine, or skip it with **E**. The CI/CD reliability report the first demo produces One agent turn from a script or a CI job, with JSON output and exit codes. ```bash theme={null} opensre --json ask "how many PRs currently have failing CI in Tracer-Cloud?" ``` See [Headless CLI](/docs/guides/headless-cli) for `--allowed-tool` and stdin prompts. Mention OpenSRE in a channel or DM it, and hand off DevOps chores to the team's agent. Demo option **D** in the shell connects Slack; see [Slack](/docs/integrations/messaging/slack). An organization admin provisions a hosted gateway at app.opensre.com; repairs and scheduled loops run there with the organization's credentials. See [Keep CI green with an agent](/docs/cicd/ci-repair-agent#the-managed-service) and [Deployment](/docs/guides/deployment#opensre-cloud-managed-gateway). Follow the [Quickstart](/docs/getting-started/quickstart) to install, sign in and run your first demo in a few minutes. ## Explore the docs * [Get started](/docs/getting-started/quickstart): install on macOS, Linux, Windows or Docker, sign in, and the enterprise setup. * [CI/CD](/docs/cicd/overview): the reliability report, fixing failing checks, the repair agent, and the organization-wide health scan. * [Using the shell](/docs/getting-started/interactive-shell-commands): every slash command, how prompts map to tools, and goals that run across turns. * [Run it anywhere](/docs/guides/deployment): the gateway daemon, headless CLI, HTTP and Python APIs, and the managed gateway. * [Platform](/docs/platform/llm-providers): LLM providers, masking, sessions, memory, scheduled loops, and closed-loop learning. * [Integrations](/docs/integrations): 60+ tools across monitoring, incidents, cloud, code, chat, databases and workflows. * [Community](/docs/guides/bi-weekly-giveaway): the showcase giveaway and the pull request review flow. ## How a turn runs Say an alert fires and you paste it in: 1. OpenSRE reads the alert and decides what kind of problem it is looking at. 2. It gathers context: who owns the service, what changed recently, what is normal. 3. It checks the tools you connected for logs, metrics and traces. 4. It keeps digging and testing ideas until it can answer. 5. It answers in the shell and sends the result where you asked: Slack, Telegram, a file.
Agent workflow
How OpenSRE Works
OpenSRE is open source. It works with the stack you already have, runs locally or hosted for a whole team, and shows how it reached every answer. The [FAQ](/docs/guides/faq) has the short answers. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.