skills/tassosgomes/my-skills/flow-qa/Gen Agent Trust Hub

flow-qa

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external documents such as PRDs (Product Requirement Documents), TechSpecs, and user stories during the 'Discover' phase. This creates a potential surface for indirect prompt injection if those documents contain malicious instructions intended to divert the agent's behavior. However, this is inherent to the QA process, and the skill mitigates this by requiring explicit user approval of the generated qa_test_plan.md in the 'Plan' phase before execution.
  • [DYNAMIC_EXECUTION]: In the 'Prepare' and 'Execute' phases, the skill materializes test cases into executable Playwright scripts. While this involves generating and running code, it is restricted to the 'ephemeral' or 'persist' modes as requested by the user and follows project-specific conventions. This behavior is standard for the skill's primary purpose of automated UI testing.
  • [CREDENTIALS_SAFE]: The skill documentation includes specific safety invariants and instructions to prevent credential leakage. It explicitly states 'Nunca grave segredos nos artefatos' (Never save secrets in artifacts) and requires the removal of tokens, passwords, and PII (Personally Identifiable Information) from test evidence.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 06:01 PM