Security audit
Security checks for vulnerabilities and agentic risk
This skill coherently converts Chinese text to audio through Zhipu AI, with the main user consideration being that submitted text is sent to an external API.
Install only if you are comfortable sending the text you provide, along with your Zhipu API authorization, to Zhipu AI's service for processing. Avoid submitting secrets, confidential business text, personal data, or regulated content unless that use is approved for your environment; install jq through your normal package-management process.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
# Configuration
API_ENDPOINT="https://open.bigmodel.cn/api/paas/v4/audio/speech"
# Get API key from environment
if [ -z "$ZHIPU_API_KEY" ]; then
echo "Error: ZHIPU_API_KEY environment variable is not set" >&2
echo "" >&2
The README instructs users to submit arbitrary text to the TTS workflow but does not clearly disclose that the content will be transmitted to Zhipu AI's external service for processing. This creates a privacy and data-handling risk because users may unknowingly send sensitive, proprietary, or regulated text to a third party.
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
# Install dependencies (if needed)
sudo apt-get install jq
# Set your API key
export ZHIPU_API_KEY="your-key-here"
The skill documents shell-based execution (bash scripts/text_to_speech.sh ...) but does not declare any permissions or allowed-tools scope. This creates an authorization gap: a host may permit shell execution without the skill explicitly constraining or disclosing that capability, increasing the chance of overbroad execution or unsafe reuse in environments with stronger tool-governance expectations.
The skill encourages users to submit arbitrary text for TTS generation but does not clearly disclose that the text is transmitted to Zhipu AI's external service. This is a privacy and data-handling risk because users may provide sensitive, proprietary, or regulated content under the mistaken assumption that processing is local.
The curl request sends the full synthesized text and an authorization bearer token to an external API endpoint. In the context of an agent skill, this is expected functionality, but it is still a genuine external transmission boundary that can expose sensitive user content to a third party if the caller is unaware or the input is not filtered.
echo "Voice: $VOICE, Speed: $SPEED, Format: $OUTPUT_FORMAT" >&2
echo "" >&2
RESPONSE=$(curl -s -X POST "$API_ENDPOINT" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $ZHIPU_API_KEY" \
-d "$PAYLOAD" \
The script transmits user-supplied text to Zhipu's remote TTS API, but it does not clearly warn the caller that the content will leave the local environment and be sent to a third party. This creates a real privacy and data-handling risk if users provide sensitive text, credentials, personal data, or proprietary content under the assumption that processing is local.
The natural-language examples and service framing are entirely Chinese-specific, which can signal a locale-specific skill behavior without an explicit opt-in or justification. Under the policy, language or locale constraints should be user-selectable or clearly documented when intentionally region-specific.
No suspicious patterns detected.