skills.rest
cure-labcure-labOfficial·6 Agent Skills Included

ArbiterOS

Runtime policy enforcement, tracing, and unsafe action interception

Enforces security policies on agent tool calls and blocks unsafe actions before side effects occur. Records authoritative traces of every plan, model call, and tool execution for full auditability. Eliminates blind trust in autonomous workflows with taint-aware dataflow checks and local deployment. Connects to existing runtimes like OpenClaw through a simple OpenAI-compatible endpoint.
175|7
↗
npx skills add cure-lab/ArbiterOS --all -g -y
Available:

Gives your AI agent the architecture, commands, and coding conventions needed to safely modify the bundled Langfuse observability stack used for trace visualization.

All Skills in This Repository (6)

Pure Emerald Level Indicators
📦 In Repo
cure-labcure-lab

vercel-react-best-practices

Optimize React and Next.js application performance with bundle size and rendering guidelines.

Official
Advanced
📦 In Repo
cure-labcure-lab

vercel-composition-patterns

Apply React component composition patterns to reduce complexity and improve maintainability.

Official
Advanced
📦 In Repo
cure-labcure-lab

backend-dev-guidelines

Guide backend development practices for Next.js, tRPC, BullMQ, and TypeScript in Langfuse's monorepo.

Official
Advanced
📦 In Repo
cure-labcure-lab

add-model-price

Add and configure LLM model pricing entries in Langfuse with regex matching.

Official
Advanced
📦 In Repo
cure-labcure-lab

skill-developer

Create and manage Claude Code skills with YAML frontmatter and trigger patterns.

Official
Advanced
📦 In Repo
cure-labcure-lab

arbiteros-instruction-metadata-dsl-writer

Write and validate ArbiterOS instruction metadata YAML DSL files.

Official
Advanced

Frequently Asked Questions

FAQPage Schema
How to install ArbiterOS?▼

Run `npx skills add cure-lab/ArbiterOS --all -g -y` in your terminal to install all skills in this suite globally.

What does ArbiterOS do?▼

It sits between your agent runtime and the model to enforce security policies, log authoritative traces, and intercept unsafe tool actions before they cause side effects.

Which agent runtimes work with ArbiterOS?▼

It currently supports OpenClaw, Nanobot, and Hermes Agent, plus any runtime that can point to an OpenAI-compatible endpoint like the default http://127.0.0.1:4000/v1.

How do I add support for a new agent?▼

Use the included DSL writer skill to create a YAML parser for the new agent's tools, validate it against the schema, run the generated tests, and register the agent name in the kernel config.

Can I visualize ArbiterOS traces and policy decisions?▼

Yes. The repository bundles an optional Langfuse stack you can start with Docker to inspect traces, policy outcomes, and runtime logs in a web UI.

Related Repositories in Software Engineering

View All in Software Engineering→
openclawopenclaw
388.4k

openclaw

Run a personal AI assistant across your devices and chat apps

Official📦 120 Skills
obraobra
280.0k

superpowers

Gives coding agents a disciplined workflow from idea to merged code

Community📦 15 Skills
facebookfacebook
248.4k

react

AI agent skills for building, testing, and porting React core

Official📦 14 Skills