insider-trading-tracker

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data sourced from external SEC Form 4 filings via the SentiSense API, which presents a surface for indirect prompt injection.
  • Ingestion points: The skill retrieves data from SKILL.md referenced endpoints: /api/v1/insider/trades/{ticker}, /api/v1/insider/activity, and /api/v1/insider/cluster-buys.
  • Boundary markers: The instructions do not specify the use of delimiters or boundary markers to isolate external data from agent instructions.
  • Capability inventory: The skill environment allows for shell command execution via npx and curl as described in the How to Run section.
  • Sanitization: The skill does not describe specific sanitization or filtering logic for fields like insiderName or securityTitle retrieved from the external API.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download and execution of a remote Node.js package using npx -y sentisense@0.47.1. This is a vendor-provided tool used to interface with the SentiSense API.
  • [COMMAND_EXECUTION]: The instructions rely on shell command execution to perform its primary tasks, including the use of npx and curl for data retrieval and authentication management.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 04:26 AM
Security Audit — agent-trust-hub — insider-trading-tracker