Caveat verdict
cursor-cloud-agents
The skill explicitly declares it reads CURSOR_API_KEY from ~/.openclaw/.env, ~/.openclaw/.env.local, .env, and ~/.cursor/config.json and caches credentials at ~/.cache/cursor-api/ without encryption; while the functionality (dispatching coding tasks to Cursor cloud) is legitimate, the broad multi-file credential harvesting and unencrypted local caching of secrets constitutes genuine 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 (7)
Possible hardcoded credential
README.md · code · API_KEY="your_cursor_api_key_here
Accesses OpenClaw config/secrets directly
SKILL.md · frontmatter · ~/.openclaw/.env
Accesses shell history/config
README.md · code · ~/.zshrc
Accesses sensitive environment variables
references/api-reference.md · code · process.env.CURSOR_API_KEY
POSTs data to external URL
references/api-reference.md · code · .post(
'https://
apt-get install — installs system packages
tests/run-tests.sh · prose · downgraded · apt-get install
Makes HTTP request to external URL
references/api-reference.md · code · fetch('https://
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class B). 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.
credential_access Is this flag fair?
Thanks — recorded.