Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward JSON2Video helper that sends user-chosen video content to the JSON2Video API, with no hidden persistence or unrelated local access found.
Install only if you are comfortable sending your video scripts, AI image prompts, voiceover text, public media URLs, and render metadata to JSON2Video and any upstream media or AI providers it uses. Keep JSON2VIDEO_API_KEY scoped to this service, avoid putting secrets or private/internal URLs in configs, and disable or adjust provider-side caching if your workflow requires tighter data retention controls.
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
"Content-Type": "application/json"
}
response = requests.post(url, headers=headers, json=payload)
response.raise_for_status()
return response.json()
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
headers = {"x-api-key": api_key}
response = requests.get(url, headers=headers)
response.raise_for_status()
return response.json()
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
def get_api_key() -> str:
"""Get API key from environment variable."""
api_key = os.environ.get("JSON2VIDEO_API_KEY")
if not api_key:
raise ValueError(
The skill documents use of environment variables, local files, and outbound network access, but it does not declare any explicit tool scope or permissions boundary. In agent environments, this can cause overbroad execution authority, making it easier for the skill to read unintended files or use secrets and network access beyond what a user expects.
The skill sends user-provided prompts, voice text, and externally hosted media URLs to the JSON2Video service, but it does not clearly warn users that this content leaves the local environment. This creates privacy and data-handling risk, especially if users include sensitive scripts, internal URLs, or copyrighted/private media without realizing they are being transmitted to a third party.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
from pathlib import Path
# API Configuration
API_BASE = "https://api.json2video.com/v2"
def get_api_key() -> str:
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
"Content-Type": "application/json"
}
response = requests.post(url, headers=headers, json=payload)
response.raise_for_status()
return response.json()
The example sets voice_id to en-US-Neural2-F, which prescribes a specific language/locale in natural-language-facing behavior. The file does not indicate that users may choose another locale or that this US English setting is only an example, so it can be read as a default locale policy without opt-in.
The examples and recommended voice formats focus on English locale voices such as en-US and en-GB, and the documentation does not mention that users may choose other languages/locales. This can be a natural-language policy issue when a skill appears to default to a specific language without opt-in or an explicit statement of flexibility.
The voice-building function sets the default language to "en", and related defaults elsewhere also assume English behavior. This can violate language/locale policy expectations because the skill imposes a specific language unless the caller explicitly overrides it.
The configuration hard-codes the voice to "en-US-EmmaMultilingualNeural", which imposes a specific language/locale choice in the skill content. There is no indication in this file that users can opt into another locale or that the U.S. English restriction is required for a region-specific purpose.
This scene again fixes the voice to "en-US-EmmaMultilingualNeural", creating the same locale policy concern. Repeating the hard-coded locale across scenes reinforces that the skill does not appear to offer language choice or explain the restriction.
No suspicious patterns detected.