extract-design-system

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches browser binaries for Chromium using the Playwright package from a well-known service.
  • [COMMAND_EXECUTION]: Executes the extract-design-system command via npx to process target URLs and manage token files as defined in SKILL.md.
  • [REMOTE_CODE_EXECUTION]: Downloads and runs the extract-design-system package from the NPM registry using npx at runtime.
  • [PROMPT_INJECTION]: Indirect prompt injection vulnerability surface. 1. Ingestion points: Processes data from arbitrary user-provided URLs (SKILL.md) and reads the resulting .extract-design-system/normalized.json file. 2. Boundary markers: The skill contains explicit instructions in SKILL.md and references/workflow.md to treat target websites and extracted output as untrusted and to avoid treating results as authoritative without review. 3. Capability inventory: Capability to execute shell commands via npx and modify project configuration files. 4. Sanitization: No specific sanitization or filtering of the scraped website content is described before the agent processes and summarizes the data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 10:42 PM
Security Audit — agent-trust-hub — extract-design-system