Resources
Everything you need to evaluate Runewrd Labs.
Documentation, source, roadmap and a plain-language glossary.
Documentation
Public Rune Box docs, quickstarts, API status, and how Guard docs are shared.
Read more →Changelog
Published Rune Box releases with dates, linked to their release notes.
Read more →Use cases
Problems Runeward Labs addresses, with exact detection IDs and stated limits.
Read more →
Rune Box on GitHub
Source code, releases and issues for the open-source governance harness.
Open on GitHub →Free assessment
What guard scan checks and how the Assessment Hub works.
Read more →Integrations
The SIEM, ticketing, chat, identity and AI harness tools Rune Guard works with.
Read more →Trust & security
Data handling, signed evidence and responsible disclosure.
Read more →Glossary
Terms we use, defined once.
- Agent harness
- The tool that runs an AI agent and mediates its access to files, shell and network, for example:
Claude Code
Cursor
- MCP
- Model Context Protocol: a standard way for agents to call external tools and data sources.
- Tool pinning
- Recording approved MCP tool definitions so later changes (a “rug pull”) are reported.
- Shadow AI
- AI tools and agents in use that security has not reviewed or approved.
- Evidence basis
- The reported source and recorded method used to link an event to an AI tool. It does not establish intent, harm or causality.
- Citadel
- A Rune Box isolated execution environment.
- Charter
- A Rune Box policy profile.
- Chronicle
- The Rune Box signed, hash-chained audit trail.
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.