Skip to content

TriggerMinds/openclaw-governance-runtime

OpenClaw Governance Runtime v1.0.0 (Provisional Template)

Overview

The OpenClaw Governance Runtime is a conceptual, standalone framework for controlled agent execution. This repository provides a suite of modular "Governance Gates" and formal schemas.

STATUS: provisional_public_runtime_template_pending_official_compatibility_reaudit

Core Principle: No Raw Execution

Every request must pass through:

  1. Normalization: telegram_command_normalizer.
  2. Compilation: runtime_command_compiler.
  3. Gating: Person-Work, Security, and Knowledge Engineering gates.
  4. Audit: Immutable decision and lifecycle records.

Repository Structure

  • governance/: Active policies for the control plane.
  • profiles/: Core roles and example work models.
  • schemas/: Formal YAML definitions for runtime objects.
  • skills/: OpenClaw-native skill directories (snake_case) with JSON metadata chosen for parser compatibility.
  • validation/: Test cases and re-audit reports.

Installation & Configuration

  • Skills: Register the skills in your openclaw.json (JSON5).
  • Configuration: See openclaw.json.concept.json5 for illustrative examples. WARNING: These are conceptual and must be adapted to your local OpenClaw installation.
  • Logs: Create a local logs/ directory (ignored by git).

Security Posture

  • Fail-Closed: Unknown states or permissions are blocked.
  • Instruction-Only: Skills are reasoning wrappers; they do not possess autonomous execution privileges.

License

Licensed under the Apache License 2.0.

About

Community governance runtime template for OpenClaw inbound commands, task-package compilation, skill security, lifecycle gates and audit-ready execution boundaries.

Topics

Resources

License

Contributing

Security policy

Stars

Watchers

Forks

Packages

 
 
 

Contributors