Back to skill

Security audit

Coze Asr

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward Coze transcription helper that uploads a user-chosen audio file to Coze for ASR, with privacy considerations but no hidden or destructive behavior found.

Install only if you are comfortable sending selected audio files to Coze's API using your COZE_API_KEY. Do not use it for confidential, regulated, or third-party recordings unless you have the required consent and have checked Coze's data handling terms.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/speech_to_text.sh (reported line 10)May include surrounding context.

sh
# Configuration
API_ENDPOINT="https://api.coze.cn/v1/audio/transcriptions"

# Get API key from environment
if [ -z "$COZE_API_KEY" ]; then
    echo "Error: COZE_API_KEY environment variable is not set" >&2
    echo "" >&2

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding

The skill documents execution of a shell script but does not declare any tool scope such as allowed shell usage or permissions. This weakens review and containment by making execution capability implicit, which can lead users or platforms to run code without clear authorization boundaries.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The documentation instructs users to transcribe local audio through Coze's API but does not clearly warn that audio content is transmitted to an external third party. This creates a privacy and data handling risk, especially if users submit sensitive recordings without informed consent or awareness of external processing.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

This skill transmits user-supplied audio to an external third-party endpoint for transcription, which creates a real data exfiltration/privacy boundary. In the context of an ASR skill this behavior is expected, but it is still security-relevant because audio may contain sensitive personal, corporate, or regulated information and is sent off-host using an API key.

Content

Scanner excerpt · scripts/speech_to_text.sh (reported line 8)May include surrounding context.

sh
set -e

# Configuration
API_ENDPOINT="https://api.coze.cn/v1/audio/transcriptions"

# Get API key from environment
if [ -z "$COZE_API_KEY" ]; then

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script sets the default transcription language to zh when the user does not provide a language argument. This imposes a specific locale choice by default rather than offering neutral behavior or explicit user opt-in, which matches the language/locale policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The description emphasizes Chinese audio transcription and the parameters set the default language to zh, while parts of the documentation and examples are also Chinese. This can impose a language/locale default without clearly obtaining user opt-in or presenting a neutral language choice up front.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.