argent-qa-flows

Pass

Audited by Gen Agent Trust Hub on Oct 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data to generate executable automation scripts.
  • Ingestion points: The skill description explicitly states it creates flows from 'test cases, tickets, or acceptance criteria', which are external, user-controlled inputs.
  • Boundary markers: There are no instructions for the agent to use delimiters or ignore embedded instructions within these external inputs.
  • Capability inventory: The skill possesses capabilities to execute local scripts via the script: fixture, run flows using flow-execute, and manage simulator environment services through tool calls.
  • Sanitization: The instructions do not prescribe any validation, escaping, or sanitization of the input data before it is translated into executable commands or scripts.
  • [DYNAMIC_EXECUTION]: The skill utilizes a script: fixture mechanism that allows for the inclusion and execution of local setup or cleanup scripts within a QA flow.
  • The instructions explain that the script: directive runs requested local setup or cleanup, which involves the dynamic execution of code within the agent's environment.
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to use CLI tools and environment management commands.
  • The agent is instructed to use argent flow run for CI environments and call stop-all-simulator-servers to reset device states, which involves interacting with the system shell and managing active processes.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 13, 2026, 07:12 PM