ClawHub

Step Asr

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward cloud transcription helper that uploads the user-selected audio file to StepFun’s ASR API and optionally saves the transcript locally.

Install only if you are comfortable sending the audio files and any prompt text you choose to StepFun for processing under your StepFun account. Avoid confidential, regulated, or third-party recordings unless StepFun’s terms and your consent requirements allow it, and keep STEPFUN_API_KEY private.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill clearly relies on environment secrets, writes output files, and sends data over the network, but it does not declare permissions to make those capabilities explicit to users or the platform. This weakens transparency and consent, especially because audio content and API credentials are involved, and can lead to users invoking a skill without understanding its access scope.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The skill transcribes local audio by sending it to the external Step ASR service, but the description and quick-start guidance do not prominently warn users that their audio leaves the local environment. Audio often contains sensitive personal, business, or regulated information, so omission of this disclosure creates a meaningful privacy and data-handling risk.

Missing User Warnings

Medium
Confidence
79% confidence
Finding
The script sends the full base64-encoded audio content to a third-party remote API, but the runtime behavior does not provide an explicit user-facing warning or consent prompt before transmission. In a skill context, users may run the tool on sensitive recordings and underestimate that audio leaves the local environment, creating privacy and compliance risk.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal