Back to skill

Security audit

Smart Trading Assistant

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent trading helper, but it gives live trade execution guidance without enough user confirmation or enforced risk limits.

Install only if you will keep this skill in analysis or paper-trading mode unless you add explicit confirmations and independent broker-side risk checks. Do not let its position-size output directly drive live orders without validating balance, risk percent, stop-loss distance, maximum size, and user approval.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/position_sizing.py:8
Finding

Position-Sizing Utility Does Not Enforce Financial Input and Risk Limits

Content
View full analysis
") sys.exit(1) try: balance = float(sys.argv[1]) risk_percent = float(sys.argv[2]) stop_loss_dist = float(sys.argv[3]) size = calculate_position_size(balance, risk_percent, stop_loss_dist) print(f"{size:.8f}") ``` ### Technical Analysis The project documentation establishes a mandatory risk range of 1% to 2% per trade, including a statement that risk must never exceed 2%. The executable implementation does not enforce that constraint. It accepts any value that Python's `float()` function can parse for the account balance, risk percentage, and stop-loss distance. Only a non-positive stop-loss distance is handled, and that condition silently returns a position size of zero. The implementation does not reject: - Risk percentages above the documented 2% maximum. - Negative risk percentages. - Zero or negative account balances. - Non-finite values such as `nan`, `inf`, and `-inf`. - Extremely small positive stop-loss distances that generate excessively large position sizes. Because IEEE-754 non-finite values are accepted by `float()`, the program can print values such as `nan` or `inf` without raising the caught `ValueError`. A downstream trading component that trusts this program's output could therefore receive an invalid or dangerously oversized order quantity. ### Attack Path 1. An operator, automation system, or attacker controlling the command-line ar ...[truncated 1441 chars]
Remediation
View remediation
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (6)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly supports automated trading analysis and trade execution, but it does not include a clear warning about real financial loss, nor does it require explicit user confirmation before any live trade execution. In a financial context, omission of these safeguards can lead users to unintentionally place real trades or over-trust the automation, causing significant monetary harm.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

Natural-language policy requires avoiding forced language or locale constraints unless the skill offers user choice or clearly justifies the restriction. This file's user-facing instructions and description are entirely in French, with no indication that another language is available or that French is required for a specific regional purpose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
97% confidence
Finding

The entire skill content is written in French, including the title and all operational guidance, with no indication that users may choose another language or locale. Per the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is explicitly documented and justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The docstring is written in French, and the script also emits a French error message later, which indicates the skill is effectively localized to a specific language without offering the user a choice. Under the policy, forcing a specific language without opt-in can be a natural-language policy violation unless the locale restriction is clearly justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

The runtime error message is presented only in French, which can impose a language choice on users without consent or configuration. This fits the policy-violation category for language/locale constraints that are not optional or justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown template is entirely written in French, including headings and field labels, which implies the skill will produce reports in a fixed language. The file does not indicate that users can choose their preferred language or that French is a documented, justified locale constraint.

Content

No source excerpt is available for this finding.