ClawHub
Back to skill

Security audit

DocStream

Security checks across malware telemetry and agentic risk

Overview

This looks like a real document-processing API helper, but users should review it because it sends documents to an external service and handles API keys too loosely.

Install only if you are comfortable sending document links, document content, request metadata, and possibly your email address to the DocStream/Voss Consulting Group endpoint. Avoid confidential, regulated, or internal documents unless that data sharing is approved. Prefer setting DOCSTREAM_API_KEY yourself, avoid DOCSTREAM_EMAIL auto-signup in logged environments, and treat any printed API key as exposed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The script automatically creates an API key when only an email address is present, which introduces account registration and credential issuance behavior beyond simple document processing. This can cause unintended disclosure of a user's email to the remote service and silently provisions a credential without explicit user consent or clear notice.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill processes user documents by sending document URLs and likely document contents to a third-party API, but the description does not warn users about this external transmission. Users may provide sensitive or regulated documents without informed consent, leading to privacy, compliance, and confidentiality risks.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The process action sends arbitrary user-supplied JSON, likely including document contents, to a remote API endpoint with no inline warning about network transfer, retention, or privacy implications. In a document-processing skill, users may provide sensitive files or text, so silent exfiltration to a third-party service creates meaningful confidentiality risk.

Missing User Warnings

Low
Confidence
89% confidence
Finding
The script consumes sensitive credentials from environment variables and can also obtain a new API key, but it does not clearly disclose this credential handling behavior to the user. This is primarily a transparency and trust issue, and becomes more serious because the newly issued key is echoed back to stderr, increasing the chance of accidental exposure in logs or terminal capture.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.