Back to skill
Skillv1.0.0

ClawScan security

binance-spot-trader · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 22, 2026, 2:24 AM
Verdict
benign
Confidence
high
Model
gpt-5-mini
Summary
The skill's code, required environment variables, and runtime instructions align with its stated purpose (an autonomous Binance spot trading bot that calls an LLM service for sentiment); nothing requested or installed appears unrelated or excessive.
Guidance
This skill appears coherent, but it performs real trades and calls an external LLM service. Before installing: (1) only use Binance API keys with withdrawals disabled and IP restrictions; prefer a sub-account with limited funds; (2) review the included scripts yourself or run them in an isolated VM/container; (3) start with paper trading or tiny amounts and monitor activity closely; (4) confirm you trust the SkillBoss provider (api.heybossai.com) since trade-related market snapshots are sent to it (the code does not send your Binance API secret).

Review Dimensions

Purpose & Capability
okName/description, required env vars (BINANCE_API_KEY, BINANCE_SECRET_KEY, SKILLBOSS_API_KEY), and the code all line up: the scripts call Binance REST endpoints and the SkillBoss LLM endpoint for sentiment. No unrelated credentials or binaries are requested.
Instruction Scope
okSKILL.md instructs running the included setup and trader scripts and creating a .env. The scripts access account balances, place orders, log trades locally, and call the SkillBoss LLM endpoint — exactly what the description promises. They do not read unrelated host files or transmit API secrets to third parties.
Install Mechanism
okNo complex install spec; setup.sh installs two PyPI packages (httpx, python-dotenv) — standard, minimal, and from a well-known registry. No arbitrary archive downloads or unusual filesystem writes.
Credentials
okRequested env vars are proportional: Binance API key/secret are needed to trade and sign requests; SKILLBOSS_API_KEY is used to call the external LLM. The skill does not request unrelated secrets or a large set of environmental access.
Persistence & Privilege
okalways is false and the skill does not attempt to modify other skills or system-wide agent settings. It runs as a user-level script and writes only local logs/trade records.