ClawHub
Back to skill

Security audit

audio-stream-upload-free

Security checks across malware telemetry and agentic risk

Overview

This skill is an audio upload helper, but it asks for broad agent authority and has overbroad instructions around credentials, local media files, and third-party uploads.

Review this before installing. Only use it with audio files you intentionally want to upload to the named streaming API, use limited-scope API keys, and avoid letting it run from broad media-editing or conversion requests. The artifact does not show malicious persistence or hidden exfiltration, but its broad triggers and raw command examples need careful user control.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (6)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The skill is presented as a narrowly scoped audio upload helper, but its capability description expands into generic create/query/modify/delete/import/export operations. That scope inflation can cause an agent or user to grant broader trust and execution authority than warranted, increasing the chance of unintended actions against local files or remote services.

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
The skill instructs use of generic shell execution and inline Python for tasks that could be implemented as a narrowly bounded upload flow. Giving an agent broad exec patterns increases the attack surface for command misuse, secret exposure, file access, and arbitrary network activity, especially when combined with user-provided variables and credentials.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger conditions are overly broad, covering video processing, audio editing, media conversion, and voice generation even though the skill is ostensibly an audio upload helper. Overbroad invocation criteria can cause the skill to activate in unrelated contexts and execute networked or shell-based workflows on sensitive files or credentials without sufficiently specific user intent.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill requests API credentials and performs file upload/network transmission without prominent warnings about sensitive data handling, third-party exposure, or destination trust. Users may inadvertently transmit private audio content and secret keys to an external domain without understanding the security implications, which is especially dangerous in an agent setting with exec capability.

External Transmission

Medium
Category
Data Exfiltration
Content
}
# ...
# 步骤1：创建音频对象
create_resp = requests.post(f'{BASE_URL}/videos/create', headers=HEADERS, json={
    'title': '我的原创音乐',
    'type': 'audio'
})
Confidence
95% confidence
Finding
requests.post(f'{BASE_URL}/videos/create', headers=HEADERS, json=

External Transmission

Medium
Category
Data Exfiltration
Content
END_POS=$((FILE_SIZE - 1))
HASH=$(md5sum /path/to/podcast.mp3 | awk '{print $1}')
# ...
curl -s -X POST "https://api-w3stream.attoaioz.cyou/api/videos/AUDIO_ID/part" \
  -H 'stream-public-key: $PUBKEY' \
  -H 'stream-secret-key: $AUTH_KEY' \
  -H "Content-Range: bytes 0-$END_POS/$FILE_SIZE" \
Confidence
95% confidence
Finding
curl -s -X POST "https://api-w3stream.attoaioz.cyou/api/videos/AUDIO_ID/part" \ -H 'stream-public-key: $PUBKEY' \ -H 'stream-secret-key: $AUTH_KEY' \ -H "Content-Range: bytes 0-$END_POS/$FILE_SI

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.