kata-code-e2e-testing

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read external project files which may contain untrusted data that could influence its behavior.
  • Ingestion points: The skill reads e2e/README.md, product specifications in docs/specs/, and existing test code in e2e/tests/ to inform test creation.
  • Boundary markers: There are no requirements for delimiters or specific markers to distinguish instructions from the data in these external files.
  • Capability inventory: The skill possesses capabilities for shell command execution (via the vp tool) and file modification for test authoring.
  • Sanitization: The instructions do not specify any validation or sanitization protocols for the ingested content.
  • [COMMAND_EXECUTION]: The skill explicitly directs the agent to execute shell commands for running and debugging tests.
  • Evidence: Instructions include running vp run e2e:ui, vp run e2e:headed, and vp check. These commands are standard for the skill's stated purpose of E2E test management.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 12:40 AM
Security Audit — agent-trust-hub — kata-code-e2e-testing