Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Quant Trading Api

v1.0.0

Professional quantitative trading API integration for Chinese securities. Supports major Chinese brokers (华泰, 银河, 广发, 中信建投) with order management, position t...

0· 70·2 current·2 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
!
Purpose & Capability
The README/SKILL.md states full integration with multiple Chinese brokers, but the included code uses mock implementations (fake tokens, simulated fills, mocked market data) rather than actual authenticated broker API calls. The BROKERS map contains real broker hostnames, but the visible code does not perform real authenticated trading requests. This is a coherence problem: a user expecting a working production connector would not get one from the provided code.
!
Instruction Scope
Instructions ask you to install Python packages and to create a config.py that contains account and password in plaintext. The code reads credentials from constructor args or environment (BROKER_ACCOUNT, BROKER_PASSWORD) but the registry metadata declares no required env vars. The skill therefore instructs/encourages insecure handling of credentials (storing them in repo files) and omits declaring required secrets.
Install Mechanism
There is no install spec; SKILL.md instructs users to pip install requests, pycryptodome, websocket-client. That is a typical, expected dependency installation for a Python skill. No unusual download URLs or archive extraction were used.
!
Credentials
The skill clearly needs brokerage credentials to be useful, but the package metadata lists no required environment variables or primary credential. The SKILL.md recommends placing account/password in config.py (a file) and the code falls back to environment variables — a mismatch. Requesting/storing plaintext broker credentials in a file is disproportionate and risky without guidance on secure storage.
Persistence & Privilege
The skill does not request always:true, does not modify other skills or system-wide configuration, and is user-invocable only. It can be autonomously invoked by the agent by default (platform default), which is expected for skills.
What to consider before installing
What to consider before installing: - Expectation mismatch: the code appears to be a mock/simulator (fake tokens, simulated fills). Do not assume it will connect to real broker APIs without further implementation or verification. - Credentials handling: SKILL.md suggests putting account/password into config.py (plaintext). That is insecure. Prefer using environment variables or a secret manager and do not commit credentials to source control. - Undeclared secrets: the skill metadata declares no required env vars or primary credential even though the code needs brokerage credentials. Treat that as a red flag — confirm how credentials are supplied and stored before use. - Audit network behavior: the BROKERS map contains real broker domains. Before running with real credentials or real money, review the entire source (including truncated parts) for any calls to unknown/third-party endpoints or logging of credentials. Run the code in a sandbox or test account first. - Dependency caution: the skill tells you to pip install pycryptodome and websocket-client. Those are common, but installing third-party packages always carries supply-chain risk; pin versions and inspect dependencies if possible. - If you need production trading: get a connector from a vetted provider or implement a well-documented authenticated client using brokers' official SDKs/APIs. If the author intends this to be a placeholder, ask for documentation that explicitly states what is mock vs. real and how to securely configure credentials. Confidence note: medium — the visible portion of the code shows mock behavior and insecure credential guidance, but the file is truncated and may contain additional network calls or hidden endpoints. Reviewing the remainder of quant_trading.py and any other files (or asking the author for confirmation) would raise confidence.

Like a lobster shell, security has layers — review code before you run it.

latestvk975sd4shjjn9g1zz3px8qxxks83c3m9

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments