Add Agent QA to review tools

[pranshuchittora]

What changed?

Adds Agent QA to the Review And Source-Trace Tools table, structured catalog, and source index. The entry describes a versionable browser/mobile QA workflow with a human review boundary.

Source traceability

• Source link(s): https://github.com/vostride/agent-qa and https://vostride.com/docs/agent-qa
• Retrieved date: 2026-08-17
• What the source supports: the FSL-1.1-ALv2 license, Node.js 24+ requirement, CLI/MCP entry points, QA workflow, and provider-backed execution model
• What the source does not support: universal compatibility with every target MCP client or provider, provider pricing, or treating automated evidence as a substitute for human review

Work fit

• Category: Review and source trace
• Audience: Developers and teams reviewing agent-produced software changes
• Work problem: Exercise browser or device workflows and retain evidence before accepting a change
• Output: Test definitions and provider-backed run evidence for configured browser or device workflows
• Human review boundary: Check test definitions, provider permissions and costs, run evidence, failures, and application side effects

Limits and affiliation

• Known limits, access requirements, or failure modes: Source-available under FSL-1.1-ALv2; Node.js 24+ is required; external provider costs may apply; client/provider coverage must be verified for the target setup
• Submitter affiliation, if any: Submitted on behalf of the Agent QA project

Checks

• I kept data/catalog.yml aligned when adding or changing a catalog entry.
• I avoided private workspace, customer, employer, credential, log, or screenshot details.
• I avoided unsupported pricing, availability, performance, adoption, or model-capability claims.
• I avoided wording that implies OpenAI affiliation, endorsement, sponsorship, certification, partnership, or approval.