trader-train

Warn

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill attempts to install the neural-trader package from the public npm registry if it is not already available in the environment.
  • Evidence: npm ls neural-trader 2>/dev/null || npm install --ignore-scripts neural-trader in SKILL.md.
  • [COMMAND_EXECUTION]: The skill executes the neural-trader command-line tool via npx using user-provided arguments, which allows for arbitrary execution of the downloaded package's logic.
  • Evidence: Multiple instances of npx neural-trader with various flags in SKILL.md.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by including user-controlled variables in shell commands without visible sanitization.
  • Ingestion points: The TICKER variable provided by the user is interpolated into npx commands.
  • Boundary markers: Absent.
  • Capability inventory: Shell command execution via Bash (specifically npx).
  • Sanitization: No sanitization or validation of the ticker input is present in the instructions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 13, 2026, 01:23 PM
Security Audit — agent-trust-hub — trader-train