groundwork-verify

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill generates JavaScript test files that utilize 'child_process.exec' to validate command-line interfaces. This allows the generated tests to execute system commands based on definitions found in the feature files.
  • [EXTERNAL_DOWNLOADS]: The skill defines a 'package.json' for the testing environment that pulls in common dependencies from the npm registry, including '@cucumber/cucumber', 'node-fetch', and 'playwright'.
  • [PROMPT_INJECTION]: The skill presents an Indirect Prompt Injection surface (Category 8) by transforming untrusted external data into executable code.
  • Ingestion points: The skill reads '.feature' files from 'docs/specs/features/' and configuration from 'docs/specs/SPEC-INTERFACE.md'.
  • Boundary markers: The skill does not implement delimiters or warnings to prevent the LLM from obeying instructions embedded within the user-provided feature descriptions.
  • Capability inventory: The generated scripts possess the ability to execute shell commands, perform HTTP requests, and control browser instances via Playwright.
  • Sanitization: There is no evidence of input validation or escaping for the strings extracted from feature files before they are interpolated into the generated test scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 02:51 PM