Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed Bocha search API wrapper that sends user-provided queries or rerank documents to Bocha and does not show hidden persistence, destructive behavior, or unrelated data access.
Install this only if you intend to use Bocha as a third-party search and reranking provider. Search terms, raw JSON fields, and reranker documents may be sent to Bocha, so avoid including private or sensitive content unless that is acceptable for your use case. Prefer an environment variable for the API key and confirm tool choice for generic web-search requests.
The skill documents executable scripts that use environment variables and make outbound network requests, but it does not declare any explicit tool scope such as permissions or allowed-tools. This creates a trust and governance gap: an agent may invoke capabilities broader than users or platform policy expect, especially when API keys and live web access are involved.
The trigger description includes broad phrases like '联网搜索', 'AI 搜索', and '事实核查/行业研报检索', which can match many ordinary user requests. Overbroad activation increases the chance the skill is selected unintentionally, causing unexpected network access, data sharing with a third-party API, or use of this skill when a narrower/local option would be safer.
The description states the skill is suited for “中文互联网检索”, which signals a language/locale constraint, but there is no explicit opt-in, user choice, or justification that this skill is intentionally region-specific. Under the policy, language or locale constraints should be optional or clearly documented as necessary.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
function endpointByMode(mode) {
switch (mode) {
case 'web': return 'https://api.bochaai.com/v1/web-search';
case 'ai': return 'https://api.bochaai.com/v1/ai-search';
case 'agent': return 'https://api.bochaai.com/v1/agent-search';
case 'rerank': return 'https://api.bochaai.com/v1/semantic-reranker';
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
function endpointByMode(mode) {
switch (mode) {
case 'web': return 'https://api.bochaai.com/v1/web-search';
case 'ai': return 'https://api.bochaai.com/v1/ai-search';
case 'agent': return 'https://api.bochaai.com/v1/agent-search';
case 'rerank': return 'https://api.bochaai.com/v1/semantic-reranker';
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
function endpointByMode(mode) {
switch (mode) {
case 'web': return 'https://api.bochaai.com/v1/web-search';
case 'ai': return 'https://api.bochaai.com/v1/ai-search';
case 'agent': return 'https://api.bochaai.com/v1/agent-search';
case 'rerank': return 'https://api.bochaai.com/v1/semantic-reranker';
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
function endpointByMode(mode) {
switch (mode) {
case 'web': return 'https://api.bochaai.com/v1/web-search';
case 'ai': return 'https://api.bochaai.com/v1/ai-search';
case 'agent': return 'https://api.bochaai.com/v1/agent-search';
case 'rerank': return 'https://api.bochaai.com/v1/semantic-reranker';
The script emits its usage and validation messages only in Chinese (用法, 查询词, 结果条数必须是整数). This imposes a specific language on all users without offering a locale choice or documenting a justified region-specific constraint, which matches the language/locale policy violation criteria.
This code presents the primary user-facing usage message only in Chinese, which is a natural-language locale constraint. The file does not indicate that the tool is region-specific or provide any opt-in or alternative language option, so it appears to enforce a specific language by default.
Detected: suspicious.dangerous_exec, suspicious.env_credential_access