argent-qa-flows

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to utilize command-line tools for executing test flows and managing the local environment. Specifically, it references argent flow run for test execution and stop-all-simulator-servers (scoped to specific devices) to ensure a clean state for consecutive test passes. These are standard operations for the described QA automation framework.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external data such as user-provided tickets, test cases, and acceptance criteria to generate executable YAML test flows. While this represents a surface for indirect instructions, the skill defines a strict 'Definition of Done' and structural contract rules that limit the influence of unstructured input to specific UI verification logic.
  • Ingestion points: Test cases, tickets, and acceptance criteria provided by the user (SKILL.md).
  • Boundary markers: The output is constrained to a structured YAML schema for Argent flows.
  • Capability inventory: The skill uses tools for UI interaction (tv-remote, keyboard), app launching (launch:), and environment management (simulator reset).
  • Sanitization: The skill emphasizes manual review of snapshots and audit tables to verify that the generated logic matches the intended contract.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 02:55 PM
Security Audit — agent-trust-hub — argent-qa-flows