Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed Chinese A-share stock-screening helper that sends user stock queries to the iWenCai API and does not show hidden persistence, local data access, destructive behavior, or deceptive instructions.
Install only if you intend to use iWenCai for Chinese A-share screening and are comfortable sending your stock-screening queries, along with an iWenCai API key, to that provider. Treat outputs as market data screening, not personalized financial advice, and prefer explicit user consent before using extra search or finance tools for enrichment.
The skill requires both environment access (IWENCAI_API_KEY) and outbound network access to the iWenCai API, but it does not declare any explicit tool scope or permission boundaries. This weakens least-privilege enforcement and makes it easier for a host agent to grant broader capabilities than intended, increasing the chance of secret exposure or unintended external calls.
The documentation explicitly authorizes using other financial or search tools when the returned data is deemed insufficient. That expands the skill beyond its stated single-provider stock-screening purpose and can cause uncontrolled data egress, prompt injection exposure from search results, or use of tools with different trust and privacy properties.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
python3 scripts/cli.py --query "今日涨停的股票" --timeout 60
### curl 示例（脱敏）
```bash
curl -X POST "https://openapi.iwencai.com/v1/query2data" \
The manifest description is broad marketing-style language that suggests the skill should be used whenever a user expresses stock-screening intent, but it does not define clear invocation boundaries, eligibility checks, or limiting conditions. In an agent environment, this can cause over-triggering in loosely related financial conversations, leading the model to invoke a third-party stock-selection capability more often than intended and potentially expose users to unsuitable financial guidance workflows.
The user-facing description is entirely in Chinese and instructs interaction through a Chinese example phrase, but the file does not indicate that the skill is region-specific or that users may choose another language. This can violate language/locale policy when a skill effectively assumes a fixed language without opt-in or justification.
The module docstring and all user-facing help/error text are written in Chinese and present the CLI as operating in that language by default. There is no option or notice offering users a language choice or documenting a justified region-specific constraint, which matches the language/locale policy violation criteria.
The description states that when users ask A股 stock screening questions, the agent 'must use this skill.' The skill content is entirely Chinese and does not indicate any user opt-in or alternative locale/language behavior, which can conflict with a policy requiring language or locale choice.