A security-focused utility to audit, review, and report on the permissions, API access, and communication channels utilized by an AI agent.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-permissions
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 openclaw-permissions 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 OpenClaw Permissions Audit Skill?
The OpenClaw Permissions Audit Skill is a vital security utility for managing Openclaw Skills, providing a transparent look into your agent's authorization landscape. It automatically parses local configuration files to produce a structured, human-readable report covering AI model providers, gateway security settings, and communication channel access.
By centralizing permission visibility, this skill helps developers maintain the principle of least privilege. It ensures that every instance of Openclaw Skills operates within its intended technical bounds, offering clear insights into enabled hooks and tool execution policies while strictly masking sensitive credentials to prevent accidental data leaks.
OpenClaw Permissions Audit Skill Use Cases
- Reviewing AI model API access profiles and fallback configurations.
- Auditing authorized users and guild IDs within communication channels like Discord.
- Verifying tool execution approval rules and per-agent capability scopes.
- Checking gateway configuration details, including port bindings and Tailscale integration status.
- Conducting security audits to ensure private keys and tokens are not exposed.
How OpenClaw Permissions Audit Skill Works
- Accesses local configuration files located in the ~/.openclaw/ directory, including openclaw.json and exec-approvals.json.
- Extracts permission dimensions such as AI model providers, channel policies, and internal hooks.
- Evaluates tool execution rules to determine which agents are allowed to run specific commands.
- Applies strict security masking rules to sensitive data, ensuring only partial IDs and existence indicators are displayed.
- Formats the aggregated data into a comprehensive Markdown report featuring tables and status icons.
OpenClaw Permissions Audit Skill Setup
To utilize the permissions audit features for your Openclaw Skills, ensure the agent has read access to the standard configuration directory. No additional dependencies are required.
# Verify your configuration files exist
ls -a ~/.openclaw/
# Ensure the identity folder is present for device audits
ls ~/.openclaw/identity/
OpenClaw Permissions Audit Skill Data Schema & Taxonomy
The skill organizes audit data into a structured taxonomy within the generated report:
| Dimension | Configuration Source | Key Metadata Tracked |
|---|---|---|
| AI Model API | openclaw.json | Provider names, Auth modes, Default models |
| Channel Access | openclaw.json | Status, Policy (allowlist/open), User counts |
| Gateway Settings | openclaw.json | Port, Bind address, Auth existence, Denied commands |
| Tool Permissions | exec-approvals.json | Profile name, Default rules, Agent-specific rules |
| Device Identity | identity/device.json | Device ID (masked), Public key presence, Operator role |
| Internal Hooks | openclaw.json | Hook names, Enabled status |
OpenClaw Permissions Audit Skill Advanced Features
- Multi-file configuration analysis covering identity, approvals, and core settings.
- Automated security masking that prevents the output of full tokens, API keys, or private PEM headers.
- Missing data handling that identifies unconfigured or unreadable files without failing the audit process.
- Detailed reporting on internal hooks such as command-logger, session-memory, and boot-md status.
SKILL.md
Loading
Related Openclaw Skills
Video Script Writer Claw
A professional AI scriptwriter designed to generate shot-by-shot video scripts, podcast outlines, and multi-episode content plans with brand-consistent styling.
GreenHelix Agent Observability Stack
A comprehensive Python-based framework for distributed tracing, metrics collection, and anomaly detection in autonomous multi-agent commerce systems.
Crypto Analyst
A comprehensive cryptocurrency market analysis and strategy toolset integrating OKX and Binance data for actionable trading intelligence.
Personalized Sleep Brainwave Intervention
A professional AI consultation agent that generates personalized neuro-modulation brainwave audio plans for sleep clinics and health practitioners.
Seedance 2.0 Storyboard Director
A professional director skill for crafting movie-grade AI video prompts and cinematic storyboards from multimodal inputs.
Text Tighten
A high-performance text optimization tool designed to sharpen writing, remove fluff, and improve structural clarity without losing original intent.
Featured*
