Testing that understands your codebase.
GoodQA scans your repo, builds guided test missions, and turns every issue into a dev-ready bug report — with recordings, repro steps, and code-level analysis attached. Keep customer relationships, agent work and team conversations in the same workspace.
Access follows your organization’s plan · No credit card required
Closing the loop that Jira, Linear, TestRail, Jam, and Loom leave open
Features
One product for the whole testing loop
From understanding your app to shipping the fix — every step is AI-assisted and built to demand minimal effort from testers while maximizing what developers receive.
AI reads your repo, then interviews you
Connect a GitHub repo and a background scan maps your stack, surfaces, and risk areas. A short AI interview fills the gaps — what the app does, who uses it, what matters most.
- Automated repo scan
- Structured app understanding
- Editable AI context
Guided testing missions with Loom-style capture
AI turns your context into step-by-step test plans. Testers run missions one step at a time while the GoodQA desktop app records screen, mic, and camera — every issue captured as it happens. The Chrome extension still works too, for anyone who prefers it.
- One step at a time
- Screen + camera recording
- Pass / fail / blocked per step
Every issue becomes a dev-ready bug
Fail a step and the composer pre-fills the canonical fields. AI guides capture — account, URL, expected vs actual, repro steps, environment — and digs into your code to deepen remediation notes.
- Repro steps + environment
- Code-level analysis
- Recording + screenshots attached
Kanban triage built for the whole loop
Drag issues across status columns, set priority and severity, assign, filter, and search. Board, list and triage views, a keyboard command palette, @mentions, linked issues, and a full activity history on every issue.
- Kanban + list views
- Keyboard-first
- Priority, severity, labels
GitHub integration, safe by default
Install the GitHub App to read your code and create issues. It can optionally open pull requests for Fix Studio's AI-drafted fixes — off by default, and only active if your organization turns it on. Export enriched bugs to GitHub Issues when you want them there.
- No repo writes by default
- One-way issue export
- Scoped installation
MCP server + a typed API for agents
Everything the UI can do, headless. Claude Code and other agents read test plans, run missions, post results, and file issues over a typed REST API and an MCP server — same tenancy and RBAC.
- Full REST API
- MCP for agents
- Scoped API keys
Beyond testing
One workspace for customers, people and agents
Keep the testing loop connected to the rest of your work. These tools are live. Access follows your organization’s settings and permissions.
Run the business
- Keep customer relationships in view
- CRM brings people, companies and deals together. Track relationships and lifecycles with notes, custom fields and webhooks.
- Try customer payments
- Stripe test mode only. Create invoices with pay links, recurring plans, refunds and bank payouts through Stripe Connect. Real-money payments are not available.
- Bring your work in and keep it moving
- Import contacts and written test plans. Use automations to run CRM workflows and inspect their outcomes.
Work with agents
- Give agents context and a way to ask
- Personas have an inbox, event stream and webhooks. Follow-ups track who you are waiting on and why. Agent knowledge and persona memory keep facts and decisions. The secrets vault and secret requests let people share credentials through a secure form.
- Run agents and follow their work
- Register agent daemons and launch runs on your server. Agent browsers give each agent a fresh browser for the sites you allow. The work coordinator tracks work and handoffs. People set priorities in the agent queue.
- Use the tools your agents already know
- Use the goodqa CLI, TypeScript SDK and agent harness plugin for Claude Code, Codex and Grok. Agent discovery at GET /v1/capabilities shows what your key can do. GET /v1/changelog lists shipped changes. Product expert MCP tools answer questions about the product and how to use it.
Work together, wherever you are
- Stay in the loop
- Link Slack for notifications and replies. Keep private text, links and files in notes to self. Share roadmap suggestions and vote on what matters to you.
- Use GoodQA on your devices
- The GoodQA mobile app brings tasks, issues, messages and notifications to iOS and Android. The desktop app runs on Windows, macOS and Linux. The Chrome extension captures screenshots and recordings with console and network context. Mobile screens that are not supported yet open in the web app.
Bring your own setup
Built, with setup required
These tools need an account, database or administrator setup before you can use them.
- Managed agent hosting
- Needs setup by GoodQA. Managed agent hosting is built, but off until GoodQA configures a hosting provider. You can launch agent runs on your own server.
- Outreach
- Needs setup. Connect your organization’s own Gmail or Outlook account. Review the contacts and messages before sending.
- Native agent mail and event bridge
- Needs setup by an administrator. Native agent mail handles agent inboxes. The event bridge turns persona messages into agent runs.
- Coordinator project
- Needs setup. Create a Coordinator project to keep agent requests, checkpoints and handoffs in GoodQA.
- Custom data
- Needs setup. An administrator must configure the organization’s PostgreSQL database before the data tools can be used.
Not publicly available
Coming soon / in private preview
These systems are built, but are not switched on for public use. There is no announced release date.
- QA Commander
- An organization’s agent with desktop, human handoff and harness choices. Built, but not switched on for public use.
- GoodQA Machines, GoodQA Connect and Browser Control
- Computer access and shared browser control. Built, but not switched on for public use.
- Discord, agent chat rooms and CSV import jobs
- Additional communication and import tools. Built, but not switched on for public use.
How it works
Three steps from repo to reproducible bug
Connect & understand
Link a repo and answer a short interview. AI builds a living understanding of your app — surfaces, flows, and where it is likely to break.
Generate & test
Generate scoped test plans, assign missions, and guide testers step-by-step with recording and screenshots — minimal effort, maximum signal.
Enrich & triage
AI enriches each issue into a reproducible bug, then your team triages it on the kanban board and ships the fix.
Agents test too.
GoodQA was built for a world where Claude Code and other agents run tests alongside people. A first-class REST API and MCP server expose everything the UI can do — read plans, run missions, post results, file and enrich issues.
Explore the developer platform# file an enriched issue as an agent
curl
https://goodqa.app/v1/projects/$ID/issues \
-H "Authorization: Bearer $KEY" \
-d '{"title":"…","type":"bug"}'
# → 201 Created · issue triaged on the board
Plans
Paid plans are not open yet. Existing organizations keep their current access.
Close the loop from test to fix.
Connect a repo, generate your first test plan, and hand your developers reproducible bugs within your configured access limits.