Lse Trading Agent

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:14 PM UTC
• Version:
  0.1.1

Overview

This LSE stock-analysis skill is coherent and disclosed, with the main caution that portfolio commands can change a local paper-portfolio file.

Install only if you are comfortable with the skill fetching market/news data from Yahoo Finance and storing a local paper portfolio in data/portfolio.json. Back up that file before using --init, --add, or --remove, and treat outputs as research rather than financial advice.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 84% confidence
• Finding: The skill documents local state mutation via `portfolio.py --init/--add/--remove`, which implies file write capability to `data/portfolio.json`, yet no corresponding permission or explicit disclosure is declared in the skill metadata. Undeclared write capability is dangerous because users and hosting platforms may assume the skill is read-only market analysis when it can persist and alter local data.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 91% confidence
• Finding: The declared purpose emphasizes screening, sentiment, risk management, and backtesting, but the documented behavior also includes persistent portfolio CRUD operations and live valuation. This mismatch is security-relevant because it obscures stateful/local-data-management behavior that changes user data, increasing the chance of unintended writes, trust abuse, or unsafe approval by reviewers who expect an analysis-only skill.

Missing User Warnings

Low

• Confidence: 88% confidence
• Finding: The portfolio commands include `--init`, `--add`, and `--remove`, but the documentation does not prominently warn that these actions modify local portfolio data on disk. Without an explicit warning and confirmation step, a user may trigger persistent changes unintentionally, which is especially risky in an agent context where commands may be run on the user's behalf.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The init operation writes a fresh portfolio structure to the target path unconditionally, which can silently destroy an existing portfolio file and erase financial records. In an agent or automation context, a mistaken invocation or path manipulation could cause irreversible data loss without any confirmation barrier.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The remove command permanently updates the portfolio by deleting a position and crediting cash with no confirmation, undo path, or dry-run mode. In trading-related tooling, destructive state changes are more dangerous because a mistaken ticker, automated invocation, or compromised agent workflow can silently corrupt position records and P&L history.

VirusTotal

64/64 vendors flagged this skill as clean.