Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Crypto Arbitrage Monitor
v1.0.0加密货币跨交易所套利监控器 - 实时监控BTC/ETH/SOL等主流币种在不同交易所的价差,发现套利机会并发送飞书/Telegram预警。支持自动计算手续费后净利润。
⭐ 0· 78·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The code uses ccxt to fetch tickers and requests to post Feishu/Telegram alerts, which matches the stated purpose. Nothing in the code requires unrelated cloud credentials. Minor mismatches: SKILL.md / README mention additional files (config.py, alerts.py) and more exchanges than the included Config, but the single monitor.py implements the functionality.
Instruction Scope
SKILL.md instructs installing python-dotenv and shows a config.py-based workflow, but the shipped package contains no config.py or alerts.py — Config is embedded in monitor.py. The docs encourage pasting webhook/bot tokens into a file (monitor.py or config), which risks secret leakage (e.g., accidental commits). The instructions do not ask to read unrelated files or system secrets, but the mismatch between docs and code is confusing and unsafe in practice.
Install Mechanism
No install spec is provided (instruction-only), and dependencies are standard PyPI packages (ccxt, requests, python-dotenv). This is typical and low-risk, but SKILL.md/README recommend installing python-dotenv even though the provided monitor.py does not use dotenv; this inconsistency should be clarified.
Credentials
No environment variables or external credentials are declared by the registry metadata. The tool requires webhook URLs and Telegram bot token/chat_id to send alerts, which is proportional to its purpose. However, the code/documentation encourage placing these secrets directly in source/config files instead of using secure env vars or secret stores, increasing risk of accidental exfiltration or repository leakage.
Persistence & Privilege
The skill does not request elevated privileges or persistent platform-wide presence (always:false). It writes a local log file ('arb_monitor.log') and runs continuously if invoked; this is consistent with a monitoring tool and not excessive.
What to consider before installing
This skill appears to implement cross‑exchange price monitoring and alerting as claimed, but there are several red flags to address before running it with real credentials:
- Documentation/code mismatch: SKILL.md and README reference files (config.py, alerts.py) and use of dotenv that are not present; the shipped monitor.py embeds Config. Review the full monitor.py to confirm what it does (the provided snippet was truncated here).
- Secret handling: The instructions lead you to paste Feishu webhook and Telegram bot token/chat_id into the config/source file. Do NOT commit secrets to source control. Prefer environment variables or a secrets manager and avoid plain-text credentials in files.
- Network behavior: The program makes outbound calls to exchange public APIs (via ccxt) and to Feishu/Telegram endpoints — this is expected, but verify the webhook URLs you provide and use test/dummy webhooks first.
- Logging: The script writes arb_monitor.log in the working directory; logs may contain sensitive timestamps/prices — store logs securely or rotate them.
- Dependency trust: It uses common PyPI packages (ccxt, requests). Install from PyPI only and consider pinning versions.
- Operational risks: The tool is monitoring-only (no trading). Do not assume it will execute trades. Be aware of race/latency and execution risks described in the README.
Actions before installing: inspect the complete monitor.py source (ensure there is no hidden exfiltration or obfuscated code), clarify where/how to store secrets safely, run first with dummy webhooks, and run in an isolated environment (container/VM) if you plan to add real credentials. If you want, I can list specific lines to inspect or suggest a safer config pattern using environment variables.Like a lobster shell, security has layers — review code before you run it.
latestvk973m51jsg6cwzpmrztrmme95583avq0
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
