quant-analyst
PassAudited by ClawScan on May 6, 2026.
Overview
This instruction-only quant skill is coherent and requests no install, code, or credentials, but live-trading use should remain human-supervised.
This skill appears benign as an instruction-only quantitative finance assistant. Before installing or using it with real systems, keep it away from broker/exchange credentials unless you intentionally grant them, require human approval for live trading or deployment, and independently verify any backtest or profitability claims.
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 strategy or trading system produced by the agent could cause financial loss if deployed without independent validation and approval.
The skill directs work toward building and testing trading systems. This is aligned with the quant-analyst purpose, but if paired with broker, exchange, or deployment tools it could have real financial impact.
Implementation approach: - Model development - Strategy coding - Backtest execution - Parameter optimization - Risk controls - Live testing
Keep the skill in advisory or development mode unless the user explicitly approves live trading, broker access, deployment, and order-routing actions.
Sensitive trading assumptions, strategy details, or historical datasets could be exposed to the agent context if the user provides them.
The skill expects retrieved context and trading data, which may include proprietary strategies or sensitive risk information. This is purpose-aligned, but users should control what context is supplied.
Query context manager for trading requirements and market focus 2. Review existing strategies, historical data, and risk parameters
Provide only the needed datasets and strategy details, and avoid including account secrets, broker credentials, or proprietary information outside the intended task.
Users could place too much confidence in generated performance claims or backtest results.
The skill includes confident profitability and performance language. It appears to be a template or aspirational example, but it could encourage over-trust if repeated without evidence.
Profitability achieved Delivery notification: "Quantitative system completed. Developed statistical arbitrage strategy with 2.3 Sharpe ratio over 10-year backtest... achieving 23% annualized returns after costs."
Treat all performance claims as unverified until independently reproduced with clean data, realistic costs, and compliance review.