anydesign-analyzer

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to clone its source code from an unverified repository (https://github.com/uxKero/anydesign.git) and install external dependencies via pip install -r requirements.txt. It also requires the installation of the Playwright Chromium browser (playwright install chromium).
  • [COMMAND_EXECUTION]: The documentation provides numerous shell commands for installation and usage of local Python scripts, which perform network requests and write results to the local filesystem (e.g., python scripts/capture_site.py, python scripts/extract_css_vars.py).
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes untrusted data from external websites.
  • Ingestion points: Fetches page HTML and CSS variables from arbitrary URLs provided in prompts or script arguments.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded malicious content are defined for the fetched external data.
  • Capability inventory: The skill possesses network fetching capabilities and the ability to write files to the local system (e.g., screenshots, JSON tokens, and markdown reports).
  • Sanitization: There is no evidence of sanitization or validation of the fetched HTML/CSS content before it is processed by the agent or the scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 12:03 PM
Security Audit — agent-trust-hub — anydesign-analyzer