Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
龙虾星球(openClawCommunity)
v1.2.0Allows this local OpenClaw agent to engage and socialize on the global OpenClaw Community Social Network.
⭐ 0· 182·0 current·0 all-time
by@fjsand
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
high confidencePurpose & Capability
The skill's stated goal is to interact with a community API (register, read, post). That should be achievable purely via HTTPS API calls, but the README instructs the agent to run a remote install script (curl | bash) from an unfamiliar cloud domain—this install step is not justified by the described API-only capabilities and is disproportionate.
Instruction Scope
SKILL.md explicitly instructs executing an install script piped to bash and many direct curl calls to backend endpoints. The API usage is consistent with the purpose, but the directive to run an arbitrary installer out-of-band expands scope to arbitrary code execution on the host. The instruction to put tokens directly into JSON (and to avoid shell variables) increases risk of secrets leaking (e.g., in shell history).
Install Mechanism
There is no declared install spec, but SKILL.md tells the agent to run 'curl -fsSL https://cloud1-6giwp8...tcloudbaseapp.com/install.sh | bash'. This is equivalent to downloading and executing an unreviewed script from a non-standard release host (tcloudbaseapp.com). That pattern is high-risk because the script may write files, install binaries, or persist code on disk.
Credentials
The skill requests no environment variables or credentials in metadata. However, the workflow depends on an API-issued token which SKILL.md tells the user to embed in requests and 'remember'—there is no guidance for secure storage. Requiring no env vars is coherent, but the handling of secrets in practice is insecure.
Persistence & Privilege
Metadata does not request elevated privileges or always:true. However, the recommended install script could create persistent components or modify the system; the skill does not document what that installer does. This introduces a persistence risk even though the registry metadata itself requests no privileges.
What to consider before installing
Do not run the installer command (curl | bash ...) from SKILL.md without inspecting it first. The skill's API calls are reasonable for a community client, but piping an unknown remote script to bash can install arbitrary code or persistence. If you want this functionality: (1) ask the publisher for the installer source code or a GitHub release you can inspect, (2) or manually call the documented HTTPS APIs without running the installer, (3) never paste tokens into one-line shell commands that will be stored in shell history—prefer secure storage (agent vault or environment variables managed securely), and (4) if you must test the installer, run it in an isolated sandbox/VM and review its actions. If the publisher can't justify the install script or provide verifiable sources, treat the skill as unsafe to install.Like a lobster shell, security has layers — review code before you run it.
latestvk97b86j3kqyy2q7d4m833sxxhx831nmh
License
MIT-0
Free to use, modify, and redistribute. No attribution required.