40Medium
The skill is vulnerable to prompt injection and path traversal, risking arbitrary file creation and persistence.
Behavior Analysis
Description verified — behavior matches claim
Claims to do
eastmoney_fin_data 妙想金融数据 skill: 本 Skill 基于**东方财富权威数据库**及**最新行情底层数据**构建，支持通过**自然语言**查询以下三类数据：
Actually does
This skill makes a POST request to `https://mkapi2.dfcfs.com/finskillshub/api/claw/query` using an API key provided via the `MX_APIKEY` environment variable. It sends a natural language query in the request body to retrieve financial data from Eastmoney's authoritative database. The skill can output the results as `.xlsx`, `.txt`, and `.json` files to `/root/.openclaw/workspace/mx_data/output/`.
Skill Info
Nameqqk000/eastmoney-fin-data
Registryclawhub
Version1.0.5
PURLpkg:clawhub/qqk000/eastmoney-fin-data@1.0.5
Stars8
Downloads1,516
Installs4
Source
SHA-256e50a6713296a...
SHA-168982d7b1dc5...
MD57cbd86f0c92f...
TLSH1d02a4ee2a06...
ssdeep192:z0oweV17...
Size8,211 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
ClawHubDocs
openclaw skills install qqk000/eastmoney-fin-dataAssessments (2)
AI Agent Traps ↗Prompt Injection Detectedmediumprompt injection
DeBERTa classifier detected prompt injection (confidence: 0.63)
63% confidence
Potential Path Traversal via Unsanitized Filenamemediumpersistence
The skill generates output files with names derived from the user's query (e.g., `mx_data_{query}.xlsx`). If the `{query}` variable is not properly sanitized, it could allow path traversal (e.g., `../`) or arbitrary file creation, potentially leading to persistence mechanisms or data overwrite.
70% confidenceline 30
输出文件: mx_data_{query}.xlsxBadge
Markdown
[](https://mondoo.com/ai-agent-security/skills/clawhub/qqk000/eastmoney-fin-data)HTML
<a href="https://mondoo.com/ai-agent-security/skills/clawhub/qqk000/eastmoney-fin-data"><img src="https://mondoo.com/ai-agent-security/api/badge/clawhub/qqk000/eastmoney-fin-data.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/clawhub/qqk000/eastmoney-fin-data.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow