ClawHub
Back to skill

Security audit

SenseAudio-ASR

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a normal SenseAudio speech-recognition helper, but its broad trigger could send generic speech-to-text requests and audio to a specific external provider without the user clearly choosing that provider.

Install only if you intend Codex to use SenseAudio for ASR work. Before sending audio, confirm the user chose SenseAudio or already has that provider in scope, and treat uploaded audio, transcripts, record queries, returned audio URLs, session IDs, trace IDs, and API keys as sensitive.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The description is broad enough to route generic speech-to-text or translation requests to this vendor-specific skill even when the user did not ask for SenseAudio. That can cause unnecessary third-party data disclosure, incorrect tool selection, and policy bypass of more appropriate local or vendor-neutral solutions.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The instruction 'Use this skill for all SenseAudio speech recognition tasks' is acceptable only within vendor scope, but in combination with the broad metadata description it reinforces over-selection and may cause the agent to prioritize this skill too aggressively. In a skill that handles sensitive audio and API-backed external transmission, overbroad routing increases the chance of unintended data exposure to the SenseAudio service.

External Transmission

Medium
Category
Data Exfiltration
Content
## Endpoint Summary

- HTTP transcription: `POST https://api.senseaudio.cn/v1/audio/transcriptions`
- WebSocket realtime: `wss://api.senseaudio.cn/ws/v1/audio/transcriptions`
- Audio analysis: `POST https://api.senseaudio.cn/v1/audio/analysis`
- Records query: `GET https://api.senseaudio.cn/v1/audio/records`
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
## Endpoint Summary

- HTTP transcription: `POST https://api.senseaudio.cn/v1/audio/transcriptions`
- WebSocket realtime: `wss://api.senseaudio.cn/ws/v1/audio/transcriptions`
- Audio analysis: `POST https://api.senseaudio.cn/v1/audio/analysis`
- Records query: `GET https://api.senseaudio.cn/v1/audio/records`
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
## Endpoint Summary

- HTTP transcription: `POST https://api.senseaudio.cn/v1/audio/transcriptions`
- WebSocket realtime: `wss://api.senseaudio.cn/ws/v1/audio/transcriptions`
- Audio analysis: `POST https://api.senseaudio.cn/v1/audio/analysis`
- Records query: `GET https://api.senseaudio.cn/v1/audio/records`
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Static analysis

No suspicious patterns detected.