Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed Chinese-language investment research workflow that uses public data sources and saves local reports, with no evidence of hidden persistence, credential use, account actions, or malicious behavior.
Before installing, understand that this skill can generate financial recommendations and position sizes, fetch public market data, and save research files locally. Treat outputs as research support, verify cited data independently, and do not run referenced workspace scripts unless you have inspected them.
The skill’s trigger phrases are broad enough that ordinary requests like '投资分析' or '研究XXX' could invoke this skill unexpectedly. Because the skill can launch multi-step workflows, fetch external data, and write research files, overbroad activation increases the chance of unintended execution and side effects without clear user intent.
The skill specifies that outputs are stored in local files under the research directory, but the description does not warn users that their prompts and generated analysis may be persisted. This can lead to unintended retention of sensitive financial queries, internal strategy notes, or confidential company targets on disk without informed consent.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Macrotrends → `https://www.macrotrends.net/stocks/charts/{TICKER}/{slug}/{metric}`
- StockAnalysis → `https://stockanalysis.com/stocks/{TICKER}/forecast/`
- NeoData → `neodata://{symbol}`（内部数据源）
- OpenAlex → `https://api.openalex.org/works?filter=...`
- World Bank → `https://api.worldbank.org/v2/country/{code}/indicator/{id}`
- 研报/新闻 → 原文URL
- **覆盖时期**：`YYYY-MM-DD ~ YYYY-MM-DD` 或 `FY2021-FY2025`
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- StockAnalysis → `https://stockanalysis.com/stocks/{TICKER}/forecast/`
- NeoData → `neodata://{symbol}`（内部数据源）
- OpenAlex → `https://api.openalex.org/works?filter=...`
- World Bank → `https://api.worldbank.org/v2/country/{code}/indicator/{id}`
- 研报/新闻 → 原文URL
- **覆盖时期**：`YYYY-MM-DD ~ YYYY-MM-DD` 或 `FY2021-FY2025`
- **验证方式**：交叉核对（至少2个独立来源）或单来源标注 `⚠ 单来源`
The skill title, instructions, and required output format are all specified in Chinese, which effectively forces a specific language for user interaction and responses. Under the policy, language constraints should either be optional for the user or explicitly justified as region-specific.
The skill is written entirely in Chinese and mandates a strict Chinese output structure, but it does not indicate that language selection is optional or that the skill is intentionally limited to a Chinese-language/regional context. This creates a natural-language policy concern because the skill effectively enforces a specific language without user opt-in.
The file prescribes the output format entirely in Chinese and provides no indication that the user can choose another language. This is a natural-language locale constraint that may violate language-choice policy because it imposes a specific language without opt-in or documented justification.
The entire skill specification mandates Chinese-language role, headings, and output format, with no indication that the user can choose another language or locale. This is a natural-language policy concern because it imposes a specific language by default rather than offering opt-in or flexibility.
The entire skill specification, required headings, and output template are written in Chinese and instruct a fixed-format Chinese report, but there is no opt-in or alternative language path for users. This creates a natural-language locale constraint that can violate organizational language-choice policies when the skill is used in broader multilingual contexts.
The file mandates a fixed output format in Chinese, including Chinese section headings and phrasing, and does not indicate that users may choose another language. This is a natural-language policy concern because it imposes a specific language/locale without documented opt-in or justification.
This markdown prompt forces a specific language/locale for the skill's operation and outputs, as shown by the role, workflow, and output templates being entirely in Chinese. The policy allows locale constraints only when they are optional, user-selected, or clearly justified as region-specific; no such opt-in or justification appears here.
Line L008 explicitly says '研究不是写报告，是形成可执行的投资判断'. Later, the same file mandates a final output titled '投资研究报告' with multiple chapters and full body content, which directly contradicts the stated principle rather than merely omitting detail.
The manifest states '不写研报，只做判断', limiting the skill to judgment rather than report writing. However, this prompt instructs generation of a structured '投资研究报告' with table of contents, full正文, metadata, and report sections, which is substantively report generation rather than only a concise decision output.
The natural-language instructions, examples, and interaction phrases all assume Chinese, including the direct invocation text and CLI examples. Under the policy, forcing a specific language without opt-in or justification is a locale/language policy violation.
The skill description, trigger conditions, workflow text, and usage examples are presented entirely in Chinese, and the example invocation assumes Chinese-language use. There is no statement that users may choose another language or locale, which can violate language-choice policy if the skill is not explicitly region-specific.
This markdown file describes the workflow and explicitly shows that output is saved to research/{公司}_{日期}.md, but it does not give the user any warning about local file creation or persistence of generated content. Because markdown files should disclose behaviors that affect user data or system state, this storage side effect should be called out more clearly.
No suspicious patterns detected.