Caveat verdict
tool-connector
The skill uses Python Playwright browser automation to capture live SSO session tokens from Slack, Outlook, Teams, Google Drive, and Grafana, then stores them in ~/.openclaw/openclaw.json; the description itself flags this as CAUTION and instructs reviewing scripts before use, indicating genuine elevated risk.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Findings (8)
Accesses OpenClaw config/secrets directly
SKILL.md · frontmatter · ~/.openclaw/openclaw.json
Instructs covert action — may act without user awareness
references/tool_connections/slack/connection-sso.md · code · silently
subprocess execution — runs system commands from Python
references/tool_connections/pagerduty/connection-api-token.md · code · subprocess.check_output(
Possible hardcoded credential
scripts/shared_utils/playwright_sso.py · prose · downgraded · TOKEN="):
tokens[
Base64 encoding/decoding
references/tool_connections/github/connection-api-token.md · code · base64-decode
Python urllib.request — network access
references/tool_connections/bitbucket-server/connection-api-token.md · code · urllib.request
Python os.environ.get — reads environment variable
references/tool_connections/pagerduty/connection-api-token.md · code · os.environ.get(
os.system/popen — direct OS command execution
scripts/shared_utils/browser.py · prose · downgraded · os.system(
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class B/D). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
Requires 2 system binaries. (1 elevated: pip).
Is this flag fair?
Thanks — recorded.