ClawHub

alarm

Security checks across malware telemetry and agentic risk

Overview

This skill matches its stated purpose as a Feishu/Lark reminder bot, but it handles chat text, voice audio, credentials, and local reminder storage that operators should protect.

Install only if you intend to run a Feishu/Lark reminder bot and are comfortable sending selected voice-message audio to the configured ASR provider. Use least-privileged Feishu app credentials, keep FEISHU_BASE_URL and SENSEAUDIO_BASE_URL set to trusted HTTPS services, protect the SQLite database, define retention/deletion rules, and give chat users appropriate notice before processing voice messages.

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 (3)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The documentation describes a flow that downloads voice messages to local storage, uploads them to an external ASR provider, persists reminder data in SQLite, and sends automated follow-up messages, but it does not mention user notice, consent, retention, or handling limits. For a skill processing potentially sensitive chat and voice content, this creates a real privacy and compliance risk because operators may deploy it without informing users that their audio is being stored and transmitted to a third party.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The code uploads a local audio file to a third-party ASR endpoint via requests.post, which transmits potentially sensitive voice content off-host. In this skill's context, that behavior is explicitly part of the feature, but the absence of any consent check, disclosure at the call site, or controls around destination trust and data handling creates a real privacy/security risk if users or operators do not realize audio leaves the local environment.

External Transmission

Medium
Category
Data Exfiltration
Content
def send_text_message(self, receive_id: str, text: str, receive_id_type: str = 'chat_id') -> dict:
        token = self.tenant_access_token()
        resp = requests.post(
            f'{self.base_url}/open-apis/im/v1/messages',
            params={'receive_id_type': receive_id_type},
            headers={
Confidence
87% confidence
Finding
requests.post( f'{self.base_url}/open-apis/im/v1/messages', params={'receive_id_type': receive_id_type}, headers={ 'Authorization': f'Bearer {token}

VirusTotal

54/54 vendors flagged this skill as clean.

View on VirusTotal