Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
lock-me-in
v1.2.0Remote browser login and session persistence for headless servers. Start an interactive browser session via a temporary public URL (cloudflared tunnel), let...
⭐ 0· 251·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill claims to provide a remote browser login proxy and the included script implements that. However the registry metadata declared no requirements while SKILL.md and the script require Playwright/Chromium, cloudflared, and Node.js. The script also hard-codes an import path ('/app/node_modules/playwright-core') and probes Playwright caches and specific filesystem paths — these implementation details are plausible for the stated purpose but are not reflected in the registry metadata and look platform-specific.
Instruction Scope
Runtime instructions and the script start a headless browser, expose a web UI, and create a temporary public cloudflared tunnel with no built-in authentication. SKILL.md tells the operator/agent to 'Send the tunnel URL to the user via their messaging channel.' Exposing an unauthenticated, short-lived public UI that accepts clicks/keyboard input and persists auth cookies is within the feature set but is a high-risk action scope (sensitive session data may be exposed if the URL leaks). The script also injects stealth/anti-detection scripts into pages — appropriate for automation but increases risk if abused.
Install Mechanism
This is an instruction-only skill with one shipped script and no install spec. No external binary is downloaded by the skill itself. SKILL.md instructs how to install cloudflared (GitHub release URL) and Playwright via npx, which are conventional but will be performed manually by the operator.
Credentials
The registry metadata listed no required env vars, but SKILL.md and the script use multiple environment variables (LOCK_ME_IN_SESSIONS_DIR, LOCK_ME_IN_CHROME_PATH, LOCK_ME_IN_PORT, OPENCLAW_PROXY_URL and in code LOCK_ME_IN_CLOUDFLARED). OPENCLAW_PROXY_URL or LOCK_ME_IN_CLOUDFLARED could carry credentials or control where traffic is proxied. The skill writes sensitive session state (storage.json containing cookies/localStorage) to disk; that is expected for the purpose but is highly sensitive and the skill does not provide enforced protection or encryption for those files.
Persistence & Privilege
The skill does not request always:true and does not appear to modify other skills or system-wide configuration. It persists session files to a sessions directory (default /data/home/.browser-sessions), which is normal for its purpose but increases blast radius if the environment is shared. The platform-default ability for the agent to invoke the skill autonomously combined with saved authenticated sessions is worth caution.
What to consider before installing
This skill does what it says (a remote headless-browser login tunnel that saves cookies/localStorage), but it has several caution points you should consider before installing or running it:
- Registry metadata omits actual runtime requirements. SKILL.md and the script require Node.js, Playwright/Chromium, and cloudflared; ensure those are installed from official sources.
- The script exposes an unauthenticated public tunnel URL (cloudflared). Treat any tunnel URL as extremely sensitive — do not share it outside a trusted channel, and prefer enabling tunnel access controls (short-lived tokens, IP whitelisting, or auth) if possible.
- The tool persists storage.json (cookies + localStorage). Those files effectively contain live authentication and should be stored with strict file permissions, encrypted at rest, and rotated/cleared when no longer needed.
- The script injects stealth anti-detection code into pages. That is commonly used for automation, but combined with persistent sessions it could be abused for automated account access. Review the code carefully and consider running it in an isolated VM/container and with an account that can be revoked (not your primary account).
- The code references a hard-coded Playwright path ('/app/node_modules/playwright-core') and probes specific home/cache paths; review and, if needed, set LOCK_ME_IN_CHROME_PATH and LOCK_ME_IN_SESSIONS_DIR to controlled locations.
- Do not set OPENCLAW_PROXY_URL or similar proxy env vars to untrusted values; a proxy can capture credentials and page content.
If you want to proceed: run the script only in an isolated environment, limit tunnel exposure (add auth or VPN), audit the entire script (including the parts not shown here), and prefer ephemeral/test credentials when using it until you are comfortable with its behavior.scripts/browser-login.mjs:419
Shell command execution detected (child_process).
scripts/browser-login.mjs:69
Environment variable access combined with network send.
Patterns worth reviewing
These patterns may indicate risky behavior. Check the VirusTotal and OpenClaw results above for context-aware analysis before installing.Like a lobster shell, security has layers — review code before you run it.
latestvk97bpxnpys110vm1qpcbpyzcsd82zhv4
License
MIT-0
Free to use, modify, and redistribute. No attribution required.