Subtitle Translator

• Outcome: Pass
• Latest audit: May 28, 2026, 9:01 AM
• Version:
  0.1.1

Overview

This skill translates user-selected subtitle files through a user-provided LLM API, with the main privacy and credential risks disclosed in its documentation.

Install only if you are comfortable sending subtitle text to the configured API provider. Use trusted HTTPS endpoints and trusted network/proxy settings, avoid command-line or plaintext-file API keys on shared systems, and do not process confidential subtitles unless you control or trust the API service.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Rogue Agent: Self-Modification, Session Persistence
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (9)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 97% confidence
• Finding: The skill documentation describes capabilities to read subtitle files, write translated output, access environment variables for API credentials, and make outbound network requests, but no declared permissions are present. This creates a real security governance gap because users and platforms cannot accurately assess or constrain what the skill can access before use.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The script sends subtitle text to a user-supplied external LLM endpoint for translation, but it provides no explicit warning, confirmation, or privacy notice that subtitle contents may leave the local environment. Subtitle files often contain sensitive or copyrighted material, so users may unintentionally disclose private data to third-party services.

External Transmission

Medium

• Category: Data Exfiltration

Content

--output video_zh.srt \
  --source-lang en \
  --target-lang zh \
  --api-url https://api.openai.com/v1 \
  --api-key sk-... \
  --model gpt-4
```

• Confidence: 93% confidence
• Finding: https://api.openai.com/

External Transmission

Medium

• Category: Data Exfiltration

Content

--source-lang en \
  --target-lang zh \
  --bilingual \
  --api-url https://api.openai.com/v1 \
  --api-key sk-...
```

• Confidence: 93% confidence
• Finding: https://api.openai.com/

External Transmission

Medium

• Category: Data Exfiltration

Content

```bash
python3 scripts/list_models.py \
  --api-url https://api.openai.com/v1 \
  --api-key sk-...
```

• Confidence: 86% confidence
• Finding: https://api.openai.com/

External Transmission

Medium

• Category: Data Exfiltration

Content

### Option 1: Environment Variables (Recommended)

```bash
export SUBTITLE_API_URL="https://api.openai.com/v1"
export SUBTITLE_API_KEY="sk-your-api-key"
export SUBTITLE_MODEL="gpt-4"

• Confidence: 90% confidence
• Finding: https://api.openai.com/

External Transmission

Medium

• Category: Data Exfiltration

Content

python3 scripts/translate_srt.py \
  -i input.srt \
  -o output.srt \
  -u https://api.openai.com/v1 \
  -k sk-your-api-key \
  -m gpt-4
```

• Confidence: 95% confidence
• Finding: https://api.openai.com/

External Transmission

Medium

• Category: Data Exfiltration

Content

```json
{
  "api_url": "https://api.openai.com/v1",
  "api_key": "sk-your-api-key",
  "model": "gpt-4",
  "batch_size": 50,

• Confidence: 94% confidence
• Finding: https://api.openai.com/

Session Persistence

Medium

• Category: Rogue Agent

Content

### Option 3: Config File (Less Secure)

Create `~/.openclaw/skills/subtitle-translator/config.json`:

```json
{

• Confidence: 92% confidence
• Finding: Create `~/.openclaw

VirusTotal

55/55 vendors flagged this skill as clean.