Alpaca Trading
PassAudited by ClawScan on May 10, 2026.
Overview
The skill is coherent for Alpaca trading, but it can use Alpaca credentials to place or cancel real orders, so users should keep strict approval controls and prefer paper trading.
Install only if you intend to let the agent interact with Alpaca. Start with paper trading, protect API keys, require explicit confirmation before every live order, and do not use --force unless you fully understand the trade being submitted.
Findings (3)
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.
A mistaken or over-broad command could submit or cancel real market orders.
The documented tool can place and cancel trades, including a force mode that bypasses confirmations; this is expected for a trading skill but can affect real money if used incorrectly.
python3 scripts/alpaca_cli.py order buy AAPL 10 ... python3 scripts/alpaca_cli.py cancel all # Cancel all open orders ... Use `--force` to skip all confirmation prompts
Use paper trading by default, require explicit per-order approval for live trading, and avoid --force or cancel-all unless the user specifically requests it.
Anyone or any agent process with access to these keys may be able to view the account and trade according to the key permissions.
The skill requires Alpaca credentials that can access account data and, when configured for live trading, authorize real trades; this is purpose-aligned but sensitive.
Requires API credentials stored in environment or config: export ALPACA_API_KEY="your-api-key" ... export ALPACA_SECRET_KEY="your-secret-key" ... ALPACA_PAPER="true" # "true" for paper, "false" for live
Use paper or least-privilege keys where possible, store keys securely, and remove live credentials when not needed.
Installing an unexpected or tampered package could affect the local environment that handles trading credentials.
The script depends on an external Python package, but the artifact set has no install spec or pinned dependency; this is normal for the integration but leaves package provenance to the user.
print("Run: pip install alpaca-py")Install alpaca-py from the official package index or a trusted environment, and consider pinning a reviewed version before using live credentials.