testing-reality-checker

Warn

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill mandates running an external script ./qa-playwright-capture.sh in the 'Mandatory Process' section. This file is not included in the skill package, representing an unverifiable dependency that could execute arbitrary commands on the system.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its automated evidence processing workflow.
  • Ingestion points: Reads data from public/qa-screenshots/test-results.json and analyzes screenshots generated by external tools.
  • Boundary markers: No delimiters or warnings are used to prevent the agent from following instructions potentially embedded in the test results.
  • Capability inventory: The agent can execute shell commands (ls, grep, cat) and the ./qa-playwright-capture.sh script.
  • Sanitization: No validation or sanitization of the test results or file content is performed before processing.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 7, 2026, 02:24 AM
Security Audit — agent-trust-hub — testing-reality-checker