Latest release: v2026.7.1Download zip
Compatibility
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The artifacts describe a ClawHub CLI for searching, installing, updating, publishing, validating, and moderating OpenClaw skills/packages; those capabilities match the code and documentation reviewed.
Instruction Scope
High-impact actions such as installing, updating, deleting, publishing, and trusted-publisher changes are explicit CLI commands, with prompts or force flags for risky overwrites and suspicious installs.
Install Mechanism
The package exposes CLI bin entries and build scripts, with no package-local postinstall hook or hidden automatic execution observed.
Credentials
The CLI uses network access to ClawHub/GitHub and writes skills, package artifacts, lockfiles, and config files locally; this is proportionate for a registry and package-management tool.
Persistence & Privilege
It persists auth tokens in the user config directory with restricted file permissions and stores local .clawhub lock/origin metadata; no background worker, auto-start persistence, or unrelated privilege escalation was found.
Scan Findings in Context
[VirusTotal telemetry] expected: Telemetry was clean with 0 malicious and 0 suspicious engine detections; this supports but does not solely determine the verdict.
[SkillSpector findings] expected: No advisory SkillSpector findings were supplied for this package release.
[Static scan findings] expected: No static-scan malicious finding was supplied in the worker context.
Assessment
Install only if you want a CLI that can manage local skills/packages and interact with the ClawHub registry. Review configured registry URLs and keep the stored ClawHub token scoped appropriately; install telemetry is documented and can be disabled with CLAWHUB_DISABLE_TELEMETRY=1.Verification
Tags
Lobster (plugin)
Adds the lobster agent tool as an optional plugin tool.
Install
openclaw plugins install @openclaw/lobster
Restart the Gateway after installing or updating the plugin.
What this is
- Lobster is a standalone workflow shell (typed JSON-first pipelines + approvals/resume).
- This plugin integrates Lobster with OpenClaw without core changes.
Enable
Because this tool can trigger side effects (via workflows), it is registered with optional: true.
Enable it in an agent allowlist:
{
"agents": {
"list": [
{
"id": "main",
"tools": {
"allow": [
"lobster" // plugin id (enables all tools from this plugin)
]
}
}
]
}
}
Using openclaw.invoke (Lobster → OpenClaw tools)
Some Lobster pipelines may include a openclaw.invoke step to call back into OpenClaw tools/plugins (for example: gog for Google Workspace, gh for GitHub, message.send, etc.).
For this to work, the OpenClaw Gateway must expose the tool bridge endpoint and the target tool must be allowed by policy:
- OpenClaw provides an HTTP endpoint:
POST /tools/invoke. - The request is gated by gateway auth (e.g.
Authorization: Bearer …when token auth is enabled). - The invoked tool is gated by tool policy (global + per-agent + provider + group policy). If the tool is not allowed, OpenClaw returns
404 Tool not available.
Allowlisting recommended
To avoid letting workflows call arbitrary tools, set a tight allowlist on the agent that will be used by openclaw.invoke.
Example (allow only a small set of tools):
{
"agents": {
"list": [
{
"id": "main",
"tools": {
"allow": ["lobster", "web_fetch", "web_search", "gog", "gh"],
"deny": ["gateway"],
},
},
],
},
}
Notes:
- If
tools.allowis omitted or empty, it behaves like "allow everything (except denied)". For a real allowlist, set a non-emptyallow. - Tool names depend on which plugins you have installed/enabled.
Security
- Runs Lobster in process via the published
@clawdbot/lobster/coreruntime. - Does not manage OAuth/tokens.
- Uses timeouts, stdout caps, and strict JSON envelope parsing.
Docs
Package
- Plugin id:
lobster - Tool:
lobster - Package:
@openclaw/lobster - Minimum OpenClaw host:
2026.4.25
