growth-ideas

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted user descriptions and includes them in the data processed for the webview.
  • Ingestion points: User input for product descriptions and stages provided in SKILL.md are passed to the agent and subsequently used to generate growth ideas.
  • Boundary markers: Absent. The skill instructions do not define clear delimiters or provide instructions for the agent to ignore potentially malicious embedded content within the user-provided descriptions.
  • Capability inventory: The skill uses the run_js tool in scripts/index.html to generate a URL for a dynamic webview and renders output in assets/webview.html.
  • Sanitization: Absent. In assets/webview.html, the innerHTML property is used to inject the idea-title and idea-desc fields into the document object model. This allows for the execution of arbitrary HTML or JavaScript if the agent-generated content includes malicious payloads originating from user input.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 08:22 AM
Security Audit — agent-trust-hub — growth-ideas