Openclaw Arbiter
Verified@AtlasPA
npx machina-cli add skill @AtlasPA/openclaw-arbiter --openclawOpenClaw Arbiter
Audits installed skills to report exactly what system resources each one accesses — network, subprocess, file I/O, environment variables, and unsafe operations.
The Problem
You install skills and trust them blindly. A skill that claims to format markdown could also open network connections, execute shell commands, or read your environment variables. Nothing reports what permissions each skill actually uses.
Commands
Full Audit
Deep audit of all installed skills with line-level findings.
python3 {baseDir}/scripts/arbiter.py audit --workspace /path/to/workspace
Audit Single Skill
python3 {baseDir}/scripts/arbiter.py audit openclaw-warden --workspace /path/to/workspace
Permission Matrix
Compact table showing permission categories per skill.
python3 {baseDir}/scripts/arbiter.py report --workspace /path/to/workspace
Quick Status
One-line summary of permission risk.
python3 {baseDir}/scripts/arbiter.py status --workspace /path/to/workspace
What It Detects
| Category | Risk | Examples |
|---|---|---|
| Serialization | CRITICAL | pickle, eval(), exec(), import |
| Subprocess | HIGH | subprocess, os.system, Popen, command substitution |
| Network | HIGH | urllib, requests, curl, wget, hardcoded URLs |
| File Write | MEDIUM | open('w'), shutil.copy, os.remove, rm |
| Environment | MEDIUM | os.environ, os.getenv, os.putenv |
| Crypto | LOW | hashlib, hmac, ssl |
| File Read | LOW | open('r'), os.walk, glob |
Exit Codes
0— Clean, all skills within normal bounds1— Elevated permissions detected (review needed)2— Critical permissions detected (action needed)
No External Dependencies
Python standard library only. No pip install. No network calls. Everything runs locally.
Cross-Platform
Works with OpenClaw, Claude Code, Cursor, and any tool using the Agent Skills specification.
Overview
OpenClaw Arbiter audits installed skills to report exactly what system resources each one accesses—network, subprocess, file I/O, environment variables, and unsafe operations. This visibility helps security teams verify permissions, enforce policies, and decide when revocation or quarantine is warranted.
How This Skill Works
It performs a structured audit of installed skills, flagging access to network, subprocess execution, file writes, environment variables, and unsafe operations. You can run a full workspace audit, an audit for a single skill, or generate a permission matrix and quick status, all via Python scripts in the local environment. It relies solely on the Python standard library, with no external dependencies.
When to Use It
- After adding new skills to a workspace to verify permissions
- During security reviews to detect elevated or dangerous access
- Before enabling skills that might execute external commands or network calls
- When onboarding third-party skills to understand their resource usage
- For ongoing risk monitoring with periodic audits and policy enforcement
Quick Start
- Step 1: Determine your workspace path for the audit
- Step 2: Run a full workspace audit: python3 {baseDir}/scripts/arbiter.py audit --workspace /path/to/workspace
- Step 3: Review results with the matrix or quick status: python3 {baseDir}/scripts/arbiter.py report --workspace /path/to/workspace; python3 {baseDir}/scripts/arbiter.py status --workspace /path/to/workspace
Best Practices
- Run a full audit after adding or updating skills in a workspace
- Regularly compare the permission matrix against established baselines
- Prioritize remediation for high/critical risks (Serialization, Subprocess, Network)
- Integrate Arbiter findings into security policies and response playbooks
- Re-audit whenever skills are updated or removed to detect changes
Example Use Cases
- A data science project adds a skill that makes outbound HTTP requests; Arbiter flags Network access and prompts a security review.
- A skill spawns subprocesses; Arbiter flags Subprocess access, triggering remediation or sandboxing.
- A CI/CD plugin writes config files during execution; Arbiter flags File Write and alerts operators.
- A third-party skill uses pickle-based deserialization; Serialization risk is detected for policy action.
- Weekly audit run identifies elevated permissions across multiple skills, informing governance decisions.
