Documentation

Find the right documentation.

Guides for Rune Guard and Rune Box: install, protect your agents and operate them, kept short.

Rune Box · public

Rune Box documentation

The guides for the open-source governance harness, moving onto this site.

Live
  • Quickstart

    Create a policy, run an allowed and a denied command, and verify the signed audit trail.

  • Install

    The signed installer, package managers and building from source.

    • Homebrew
    • pip
    • npm
  • Concepts

    The product vocabulary and the governed path: Citadels, Charters, policy and approvals.

  • Policies (Charters)

    Anatomy of a Charter: rules, runtime hardening, network and budget settings.

  • CLI reference

    Every runeward command and flag.

  • REST API

    The governed HTTP API behind the dashboard, SDKs and MCP.

  • Adapters

    Framework adapters, plus LangChain.js.

    • LangChain
    • CrewAI
    • LlamaIndex
    • OpenAI Agents
    • Strands
    • Vercel AI SDK
  • Agent harnessing

    Routing a parent agent and its subagents through Runeward Labs.

  • Agent groups

    Cohorts of peer workers that iterate on one app or fan out independent pieces.

  • Security model

    What Runeward Labs provides, what is in scope for reports, and what operators own.

  • Why governance

    Why rules belong outside the agent rather than in its prompt.

  • Observability

    Metrics, a dashboard and alert rules.

    • Grafana

Rune Gurd Live

Rune Gurd documentation

Install the Guard agent, run guard scan, put hooks on your coding agents and operate Guard Cloud and the Console. Each guide says what is built today, and Known limits lists what is not.

Quickstarts

Two ways to try Runewrd Labs products today.

Live

Prove governance with Rune Box

Needs a running container engine:

  • Docker
  • OrbStack
  • Podman
brew install Runewardd/tap/runeward
runeward quickstart
runeward evidence export quickstart -o run.json
runeward evidence verify run.json
Live

Scan one machine with Gurd

A read-only report of risky agent settings, rules and skills. No network, no file changes.

guard scan
guard scan --discover
guard mcp-inventory --discover
What the scan checks →

API reference

APIs and their publication status.

Where each Runeward Labs API is documented
APIWhere it is documentedStatus
Rune Box REST APIREST API guide (moving to this site)Live
Rune Box Python and TypeScript SDKsAdapters guide (moving to this site)Live
Guard Console APIShared with early-access customersLive
Versioned public API reference and docs pipeline for this siteNot published yetPreview

Your agents already have access. Ward off the risks.

Join Guard early access for your team, or run the free local scan today. Agents in automation too? Add the open-source Rune Box.