Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward Naver Shopping search helper that uses documented Naver API credentials and does not show hidden persistence, destructive behavior, or unrelated data access.
Install this only if you intend to use Naver Shopping's API and are comfortable storing Naver API credentials in the documented environment variables or credential files. Search queries will be sent to Naver to retrieve product results.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
def _resolve_credentials():
skill_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
load_env_file(os.path.join(skill_dir, '.env'))
load_env_file(os.path.join(os.path.expanduser('~'), '.openclaw', 'credentials', 'naver-shopping.env'))
client_id = (
The skill invokes a Python script that uses network access and reads credentials from environment variables and local files, but the manifest does not declare any tool scope or allowed-tools. This creates an authorization gap: an agent or reviewer cannot easily tell that the skill needs env, file-read, and network capabilities, increasing the chance of unintended secret access or external data exfiltration.
This code automatically loads credentials from the skill-local .env file and from ~/.openclaw/credentials/naver-shopping.env, but there is no user-facing warning, prompt, or explanatory comment indicating that local credential files will be accessed. Accessing secrets or credential stores is a safety-relevant operation under the rule and should be disclosed.
The script sends the user-provided search query to Naver's external API and attaches client credentials in request headers, but there is no visible user warning or disclosure in code output/help about this transmission. Network calls that transmit user or system data should be clearly disclosed unless the behavior is explicitly communicated.
The error message is hardcoded in Korean, which imposes a specific language on users without opt-in or any fallback. This is a natural-language policy issue because the skill does not offer a language choice or document a justified locale restriction.