P2p Lending Data
Security checks across malware telemetry and agentic risk
Overview
The skill is advertised as Frappe Lending test support, but its instructions also steer the agent toward quant trading/backtesting and possible purchase or broker-account activity without clear limits.
Review this skill carefully before installing. It may be useful for finance experiments, but it is not coherently limited to Frappe Lending tests. Treat any broker, paid data, crypto, purchase, or live-trading path as high risk; use dry-run/backtest mode, isolated environments, and explicit confirmations only.
VirusTotal
66/66 vendors flagged this skill as clean.
Risk analysis
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 user expecting loan-module test help could instead get instructions for financial market strategy and trading workflows.
The same artifact presents the skill as Frappe Lending loan testing while also defining a market/trading pipeline, which could mislead users about what the skill will guide the agent to do.
description: 验证 Frappe Lending 贷款模块核心流程... Pipeline `data_collection -> data_storage -> factor_computation -> target_selection -> trading_execution -> visualization`
The publisher should split the lending-test and quant-trading content, rename the skill, or clearly disclose the trading behavior before invocation.
If connected to tools or accounts that can trade or purchase, the agent may receive overly broad guidance for financial actions.
These instructions describe order execution and sizing behavior, but the artifacts do not clearly restrict them to simulation, dry-run use, or explicit user approval.
`SL-01` | Execute sell orders before buy orders in every trading cycle ... `TradingSignal` MUST have EXACTLY ONE of: position_pct, order_money, order_amount
Require explicit user confirmation for any order-like action, default to backtesting only, and document hard limits for live trading or purchases.
The agent could prompt for or use financial provider or broker account access that was not expected from the skill description.
The skill contemplates paid provider and broker-backed workflows even though the registry declares no primary credential and the stated purpose is lending-module testing.
Data source / provider: eastmoney (free, no account), joinquant (account+paid), baostock (free, good history), akshare, or qmt (broker)?
Do not provide broker or paid-provider credentials unless the skill is explicitly scoped for that account, preferably read-only and separate from live trading authority.
Using the quant workflow may install third-party packages and create local data directories not obvious from the registry metadata.
The registry says there is no install spec, but the reference docs include package installation and initialization commands. This is not automatically malicious, but users should notice the undeclared setup dependency.
on_fail: Run: python3 -m pip install zvt then re-run: python3 -m zvt.init_dirs
Install dependencies only after reviewing them, use an isolated environment, and ask the publisher to declare setup requirements in metadata.