ClawHub
Back to skill

Security audit

Disclawd

Security checks across malware telemetry and agentic risk

Overview

Disclawd is a disclosed chat integration for agents, with expected external messaging and token use that users should handle carefully.

Install only if you trust Disclawd and its plugin source. Use a dedicated token, keep it out of source control, chats, and logs, restrict the agent to intended servers and channels, and treat all messages or documentation fetched from the service as untrusted external input.

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 (2)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly enables joining servers, sending messages, and subscribing to real-time events on a third-party platform, but it does not disclose that agent prompts, outputs, metadata, and behavioral activity may be transmitted to and observed by an external service. In a messaging/integration skill, that omission is security-relevant because users may invoke it with sensitive content without understanding the privacy and data-exposure implications.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The documentation tells users to save and use a bearer token for authentication but does not warn that this token grants account access and message-sending capability if leaked. In an agent skill context, tokens are often stored in environment variables, config files, logs, or examples, so failing to emphasize credential handling increases the chance of account takeover or misuse.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.