ClawAudit verdict
OpenClaw Setup
modelwise-openclaw-setup
Accesses credentials AND makes external network calls
OpenClaw Setup scores 20/100 (Dangerous). It declares no permissions. 2 undeclared capabilities detected — the skill does more than its permissions suggest. 4 critical pattern matches in code.
What it does
These are capability combinations: each listed behavior occurs in the skill, but ClawAudit detects co-occurrence — it does not verify that one flows into another. Read the code to confirm a live chain.
Accesses credentials AND makes external network calls — potential credential theft
LLM02 · ASI03
Permission integrity
network_out
credential_access
package_install
Findings (11)
Pipe-to-shell pattern (curl | sh) — supply chain attack vector
SKILL.md · code · curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.1/install.sh | bash
Pipe to bash — executes piped content as shell commands
SKILL.md · code · | bash
Possible hardcoded credential
SKILL.md · code · Token: "YOUR_BOT_TOKEN
Recursive delete from root or home — destructive command
SKILL.md · code · rm -rf ~
Accesses shell history/config
SKILL.md · code · ~/.zshrc
Accesses OpenClaw config/secrets directly
SKILL.md · code · ~/.openclaw/openclaw.json
Downloads executables from external URLs
SKILL.md · code · install from https://
References sudo — requests elevated privileges
SKILL.md · code · sudo
HTTP request to bare IP address — common in malicious payloads
references/configuration-reference.md · code · http://127.0.0.1
Changes file ownership
SKILL.md · code · chown
Opens WebSocket connection
references/configuration-reference.md · code · WebSocket
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A/C). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
network_outpackage_installcredential_access Thanks — recorded.