E1QA
Functional Testing Agent
Generates and runs unit/integration/E2E tests (pytest, Playwright) against acceptance criteria.
How it runs
- What it reads
- PRD acceptance criteria, code changes.
- What it produces
- Test suites, pass/fail reports, coverage metrics.
- When it runs
- On every PR/commit (CI gate).
The same guardrails as every other agent
Functional Testing Agent is governed exactly like the rest of the fleet. Its rule file is versioned, never edited in place. Its tool list is a declaration, not a grant — it can only call a tool a gateway policy has explicitly allowed in that project. Every tool call it makes is policy-checked and written to the audit log before it leaves the platform, and a project's cost cap is checked before a run starts, not reconciled after it.
More in QA
- E2Code Review AgentStatic analysis, style/quality gate, and architectural-consistency check before human review.Read the product page
- E3Security & Penetration Testing AgentRuns vulnerability scans (SAST/DAST), dependency audits, and simulated attack scenarios.Read the product page
- E4Performance/Load Testing AgentSimulates production load, measures latency/throughput, identifies bottlenecks.Read the product page
- E5UAT / Stakeholder Acceptance AgentCoordinates user acceptance testing with actual business stakeholders, packages test scenarios from acceptance criteria, and collects sign-off before the release gate.Read the product page
- 188Database Testing Agentindependently verifies schema migrations, query performance, and data integrity before and after they ship; distinct from the Database Development Agent, which self-reports on its own migrations and queries, and from Performance/Load Testing, which measures application-level throughput, not database-specific query plans or referential integrity.Read the product page
- 189Visual Regression Testing Agentcaptures screenshots of rendered UI across breakpoints and diffs them against an approved visual baseline to catch unintended visual regressions — layout shifts, broken CSS, misaligned components; distinct from Functional Testing (behavioral correctness against acceptance criteria) and the Accessibility Agent (semantic/WCAG compliance), neither of which verifies pixel-level rendered appearance.Read the product page
- 190Responsive Design Testing Agenttests rendered layouts across a continuous range of viewport widths, not just discrete breakpoints, for responsive correctness: content overflow, touch-target sizing, text truncation, and broken mid-range layouts between defined breakpoints; distinct from Visual Regression Testing, which diffs discrete breakpoint snapshots against a prior baseline and cannot catch a responsive flaw present since a screen's own first version, nor a bug that only manifests between listed breakpoints.Read the product page
- 191Source Code Auditing Agentperforms a periodic, whole-codebase audit, not scoped to a single PR diff, for hardcoded secrets, missing license/copyright headers, and code whose own structure closely resembles restrictively-licensed external sources; distinct from Code Review Agent, which gates new PR diffs against house style, and Security & Penetration Testing, which scans for known vulnerability patterns and CVEs rather than provenance or license risk.Read the product page