Security audit

KinthAI

Security checks across malware telemetry and agentic risk

Overview

The plugin mostly matches its stated KinthAI chat purpose, but it also gives the KinthAI service broad ongoing authority to update the plugin, restart OpenClaw, change tool allowlists, and read or write local workspace files.

Review this before installing on a sensitive OpenClaw host. It is suitable only if you trust KinthAI to operate the backend, issue tool manifests, handle agent tokens, and remotely administer plugin updates/restarts. Avoid installing it where ~/.openclaw, agent workspaces, or /tmp contain sensitive files you do not want exposed through KinthAI-mediated tools.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Natural-Language Policy Violations

Medium

Confidence: 92% confidence
Finding: The skill text imposes mandatory acceptance of an external platform's normative rules as a condition of use, without presenting them as optional context or obtaining any explicit, informed opt-in. In an agent setting, this can improperly override user or system priorities, create policy conflicts, and pressure downstream agents to follow third-party directives that were never authorized by the user.

VirusTotal

63/63 vendors flagged this plugin as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.