- name
- skill-sandbox
- description
- Test untrusted skills in an isolated environment before installing. Monitors network access, filesystem writes, environment variable reads, and subprocess calls. Run any skill safely without risking your agent's data or credentials.
- user-invocable
- true
- metadata
- {"openclaw": {"emoji": "🏖️", "os": ["darwin", "linux"], "requires": {"bins": ["python3"]}}}
Skill Sandbox
Run untrusted skills in a monitored environment. See exactly what they do before giving them access to your real system.
Why This Exists
ClawHub has hundreds of skills. Some are malicious. Even after scanning with arc-skill-scanner, you can't catch everything with static analysis. The sandbox lets you run a skill's scripts and observe their behavior at runtime — what network calls they make, what files they access, what environment variables they read.
Commands
Sandbox a skill directory
python3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/Run a specific script in sandbox
python3 {baseDir}/scripts/sandbox.py run --script ~/.openclaw/skills/some-skill/scripts/main.pyRun with network monitoring
python3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/ --monitor-networkRun with fake environment variables
python3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/ --fake-envRun with a time limit
python3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/ --timeout 30Generate a safety report
python3 {baseDir}/scripts/sandbox.py report --path ~/.openclaw/skills/some-skill/What It Monitors
Filesystem Access
- Files opened (read/write)
- Directories created
- File deletions
- Permission changes
Environment Variables
- Which env vars are read
- Whether sensitive keys are accessed (API keys, tokens, passwords)
- Option to inject fake values to see what the skill does with them
Network Activity
- Outbound HTTP/HTTPS requests (URLs, methods, payloads)
- DNS lookups
- Socket connections
- FTP, SMTP, and other protocols
Process Execution
- Subprocess calls
- Shell commands
- Dynamic imports
Safety Modes
- observe (default) — Run the skill and log everything it does. No restrictions.
- restricted — Block network access and filesystem writes outside a temp directory.
- honeypot — Provide fake credentials and endpoints to see if the skill tries to exfiltrate.
Output
The sandbox produces a JSON report with:
- All filesystem operations (reads, writes, deletes)
- All environment variable accesses
- All network connections attempted
- All subprocess calls
- Warnings for suspicious patterns
- A safety verdict (SAFE / SUSPICIOUS / DANGEROUS)
Integration
Combine with the workflow orchestrator for automated pre-install checks:
scan skill → sandbox run → review report → install if safe → audit logLimitations
- Python skills only (JavaScript/shell support planned)
- Cannot catch all evasion techniques (obfuscated or delayed execution)
- Network monitoring requires the skill to use standard Python libraries
- Not a true OS-level sandbox (use Docker for that level of isolation)