grading-frontend

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute utility packages including @axe-core/cli, lighthouse, playwright, and ajv-cli. These are well-known services and standard industry tools used for auditing web applications.
  • [COMMAND_EXECUTION]: The skill executes local shell scripts provided within its directory (probes/probe-screenshots.sh, probes/probe-performance.sh, probes/probe-a11y.sh). It also uses shell utilities like grep, find, wc, and rg to perform static analysis on the student's project files.
  • [INDIRECT_PROMPT_INJECTION]: As a grading tool, the skill naturally processes untrusted student project files and audits local dev servers.
  • Ingestion points: Project source files (e.g., package.json, .tsx, .css) and the output of the local dev server at http://localhost:3000.
  • Boundary markers: No explicit markers are used to isolate untrusted content from the instructions.
  • Capability inventory: The skill can execute shell commands, read and write local files, and make network requests via automated browsers to audit the application.
  • Sanitization: The skill relies on standard output from auditing tools and does not explicitly sanitize the source content of the projects being graded.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 10:54 PM