qa-flow-report

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the execution of a local TypeScript file (render-qa-flow-report.ts) using the Bun runtime to generate the final HTML report. This command is executed on local files within the skill's own directory structure as part of its core functionality.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it ingests untrusted data from the browser environment (console logs and network request metadata).
  • Ingestion points: Browser console messages and network request URLs captured via mcp__chrome-devtools__list_console_messages and mcp__chrome-devtools__list_network_requests in SKILL.md.
  • Boundary markers: The agent is instructed to record these as data in a structured manifest.json rather than interpreting them as commands.
  • Capability inventory: The skill uses Chrome DevTools tools to interact with web pages and read local files to generate reports.
  • Sanitization: The rendering script render-qa-flow-report.ts implements a comprehensive escapeHtml function that sanitizes all strings before they are interpolated into the HTML template, effectively mitigating XSS risks from malicious site content.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 02:45 PM
Security Audit — agent-trust-hub — qa-flow-report