# Xerg

## Docs

- [Audit](https://xerg.ai/docs/audit.md): Run a local Xerg audit, choose output formats, control snapshot persistence, and understand the core report.
- [Authentication and push](https://xerg.ai/docs/auth-and-push.md): Use connect for the guided hosted path, push explicit audit summaries when you want to, and keep local auditing separate from paid hosted features.
- [CI gates](https://xerg.ai/docs/ci-gates.md): Fail CI when confirmed waste exceeds a threshold and optionally push the audit to the Xerg API.
- [CLI cheat sheet](https://xerg.ai/docs/cli-cheat-sheet.md): A sequenced Xerg CLI reference covering init-first onboarding, direct audit flows, hosted follow-up, remote OpenClaw audits, and CI gates.
- [Compare](https://xerg.ai/docs/compare.md): Understand how Xerg chooses a baseline and how to make before and after comparisons meaningful.
- [Doctor](https://xerg.ai/docs/doctor.md): Check local, SSH, or Railway audit readiness before you run a full Xerg audit.
- [Findings and recommendations](https://xerg.ai/docs/findings-and-recommendations.md): Understand Xerg's finding kinds, what is confirmed waste, and what the v2 recommendation objects mean.
- [Hosted MCP (Pro)](https://xerg.ai/docs/hosted-mcp.md): Configure Cursor, Claude Code, Codex, or another client for Xerg's hosted MCP server after your first pushed audit.
- [How it works](https://xerg.ai/docs/how-it-works.md): Understand Xerg's local-first audit flow, local snapshot storage, remote pull model, and privacy boundaries.
- [Xerg](https://xerg.ai/docs/index.md): Start with local OpenClaw or Hermes audits, compare fixes, use remote OpenClaw audits when needed, and add hosted follow-up only if you want it.
- [Init](https://xerg.ai/docs/init.md): Use xerg init for the first local audit, understand what it detects, and know when to switch to direct commands.
- [Installation](https://xerg.ai/docs/installation.md): Install and run the Xerg CLI, know the supported runtimes, and start with init or direct audit commands.
- [JSON output](https://xerg.ai/docs/json-output.md): Use Xerg's machine-readable summary and push payload formats for automation.
- [Metrics](https://xerg.ai/docs/metrics.md): Learn what Xerg's top-level dollar metrics mean and how before and after deltas are computed.
- [Quickstart](https://xerg.ai/docs/quickstart.md): Use the init-first local path, compare the next run, and only opt into hosted follow-up if you want it.
- [Remote audits](https://xerg.ai/docs/remote-audits.md): Audit OpenClaw data on SSH hosts or Railway services from your local machine.
- [Skill bundle](https://xerg.ai/docs/skill-bundle.md): Find the bundled Xerg agent skill in the npm package and understand what it is for.
- [Troubleshooting](https://xerg.ai/docs/troubleshooting.md): Solve the most common Xerg install, audit, compare, remote, and push issues.
- [Waste taxonomy](https://xerg.ai/docs/waste-taxonomy.md): Understand Xerg's finding kinds, confidence levels, and the difference between confirmed waste and directional opportunities.

## OpenAPI Specs

- [openapi](https://xerg.ai/docs/api-reference/openapi.json)

## Optional

- [npm](https://www.npmjs.com/package/@xerg/cli)
- [Pilot](https://xerg.ai/pilot)