AI AGENTS
BoringOS
by Hebbs
Overview
Open-source agent framework and shell for running coding agents as CLI subprocesses with tasks, workflows, memory, approvals, budgets, and audit trails.
Details
BoringOS is an open-source agent framework and shell from Hebbs for operating coding agents from the command line. The supplied website, documentation, and GitHub metadata describe BoringOS as running coding agents as CLI subprocesses and adding structure around tasks, workflows, memory, approvals, budgets, and audit trails. Its module quickstart describes creating a Module manifest with skills and tools, registering the module on a BoringOS host, and exposing callable agent capabilities. The Tool spec describes module-registered operations callable by agents through POST /api/tools/<module-id>.<tool-name>, with validation and auditing.
When to Use
Use BoringOS when you want an open-source shell/framework for running coding agents as CLI subprocesses. Use it when coding-agent work needs task and workflow structure plus memory approvals budgets and audit trails. Use it when you want to expose module-defined skills and tools as callable capabilities on a BoringOS host.
Getting Started
- Visit the BoringOS website for the project overview: https://www.boringos.dev/
- Read the documentation: https://www.boringos.dev/docs
- Review the GitHub repository: https://github.com/hebbs-ai/boringos
- Follow the module quickstart to scaffold or hand-roll a Module manifest
- register it on a BoringOS host
- and expose callable capabilities.
- Review TOOLS.md for the Tool spec and how module-registered operations are called by agents.
Key Features
- •Runs coding agents as CLI subprocesses
- •Task and workflow structure
- •Memory support
- •Approvals
- •budgets
- •and audit trails
- •Module manifests containing skills and tools
- •Module registration on a BoringOS host
- •Tool calls via POST /api/tools/<module-id>.<tool-name> with validation and auditing
Capabilities
- •CLI coding-agent orchestration
- •Agent framework and shell
- •Module-based skill and tool exposure
- •Task and workflow management
- •Audited tool invocation
Last updated Jun 10, 2026