Mx Finance Data 1.0.8
Security checks across malware telemetry and agentic risk
Overview
This appears to be a coherent finance-data query skill, but it uses an Eastmoney API key, sends queries to Eastmoney, and installs unpinned Python packages.
Install only if you trust the publisher and Eastmoney service, use a revocable EM_API_KEY, and avoid including confidential investment or client information in natural-language queries.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Installing the skill may fetch the latest available versions of these packages, so behavior could change if dependencies change upstream.
The skill asks the user/platform to install unpinned Python dependencies. This is common for a Python data tool, but dependency versions are not fixed.
pip3 install httpx pandas openpyxl --user
Install in a virtual environment where possible and prefer pinned dependency versions from a trusted package index.
The key may identify the user to Eastmoney and may consume quota or carry account permissions for that service.
The script reads an Eastmoney API key from the environment and places it in the API request context as the user identity.
EM_API_KEY = os.environ.get("EM_API_KEY", "") ... "userId": EM_API_KEYUse a scoped, revocable API key, keep it out of prompts/logs/output files, and rotate it if it may have been exposed.
Financial entities, strategies, or research questions included in the query may be visible to the remote provider.
The user’s natural-language financial query is sent to an Eastmoney remote API endpoint, which is expected for this provider-backed skill.
DEFAULT_SEARCH_API_URL = "https://ai-saas.eastmoney.com/proxy/b/mcp/tool/searchData" ... "query": query
Avoid putting confidential client, trading, or proprietary strategy information into queries unless sharing it with Eastmoney is acceptable.
