A mandatory security wrapper for strict-mode execution, evidence logging, and policy enforcement in autonomous operations.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install governance-wrapper
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install governance-wrapper using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Governance Wrapper?
The Governance Wrapper is a critical security layer designed for managing autonomous agent behavior within strict environments. It serves as a mandatory execution gatekeeper, ensuring that every interaction follows predefined safety protocols and organizational policies. By utilizing this tool to wrap Openclaw Skills, developers can ensure that agents remain within operational bounds, using only approved models and network paths while maintaining a tamper-proof audit trail of every action taken during an autonomous session.
Governance Wrapper Use Cases
- Enforcing strict-mode execution for production-level autonomous agents
- Generating immutable evidence logs for regulatory compliance and security auditing
- Restricting autonomous agent network access to approved allowlists for outbound HTTP traffic
- Managing resource allocation for complex workflows using subagent semaphore controls
- Ensuring consistent and deterministic model behavior by locking execution to specific LLM versions
How Governance Wrapper Works
- The wrapper is invoked with mandatory parameters including the requested skill, system prompt, and input context.
- It validates the request against the tool surface manifest and network allowlist before proceeding.
- The system locks the execution to the opencode/big-pickle model with a temperature of 0.0 for maximum consistency.
- During execution, the wrapper monitors all activities in real-time to ensure no security or policy violations occur.
- It generates a hash-chained, append-only evidence log (execution-evidence.v1) to record the entire lifecycle of the operation.
- Any detected policy violation results in an immediate block of the execution with recorded evidence of the attempt.
Governance Wrapper Setup
To deploy the wrapper within your workspace, ensure the governance script is correctly located. You can then execute your Openclaw Skills through this secure wrapper using the following command structure:
python3 /home/openclaw/.openclaw/workspace/governance/governance_wrapper.py \
--requested-skill <skill_name> \
--system-prompt "your_system_prompt" \
--input-context "your_input_context"
Governance Wrapper Data Schema & Taxonomy
The Governance Wrapper produces structured execution evidence and enforces the following technical specifications:
| Component | Specification |
|---|---|
| Model Lock | opencode/big-pickle (No Fallback) |
| Parameter Control | Temperature 0.0 |
| Evidence Format | execution-evidence.v1 (Hash-chained) |
| Concurrency | Managed via maxConcurrentSubagents semaphore |
| Network Security | Outbound HTTP allowlist enforcement |
Governance Wrapper Advanced Features
- Subagent semaphore management to strictly control the maxConcurrentSubagents limit
- Mandatory hash-chained append-only logging to prevent evidence tampering or deletion
- Strict enforcement of model parameters to ensure deterministic outcomes for all Openclaw Skills
- Integrated skill allowlist enforcement based on the tool surface manifest
SKILL.md
Loading
Related Openclaw Skills
OctoFlow
A zero-dependency, GPU-native programming language that converts natural language descriptions into high-performance Vulkan compute programs.
Wisdom & Accountability Coach
A sophisticated AI mentor that uses longitudinal memory and philosophical frameworks to guide personal development and maintain compassionate accountability.
Project Management Guru (ADHD-Specialized)
A specialized project management agent that protects hyperfocus and minimizes context-switching costs for ADHD-driven engineering workflows.
Jungian Psychologist
An expert AI skill providing structured guidance in Jungian analytical psychology, shadow work, and archetypal analysis.
DeShell
A high-performance proxy skill that converts web pages and search results into LLM-friendly Markdown to optimize token consumption.
Distil
Distil enables AI agents to search the web and fetch web pages as clean, LLM-friendly Markdown to drastically reduce token consumption.
Featured*
