Testsigma
Agentic test automation platform whose AI agents generate, run and self-heal tests for web, mobile, API and Salesforce apps.
Testsigma is a test automation platform for QA teams that relies on AI agents. Testsigma connects to tools such as Jira and GitHub, works out what a code change affects, writes test cases in plain English, runs them on real browsers and devices, repairs broken locators and reports a release confidence score. Pricing is quoted individually.
Best for QA teams that want broader automated coverage without writing scripts and need a clear go or no-go signal before each release.
- Platforms
- Web
- Pricing model
- Contact sales
- Free trial
- Yes
- Added
Last verified: . Information taken from testsigma.com.
What is Testsigma?
Testsigma is a cloud platform for automated software testing built around AI agents, which the company calls Atto. On each commit the agents map the change to the related sprint, stories and specifications, show what is and is not covered, and draft missing test cases as editable plain-English steps grouped into smoke, feature, regression and deep regression plans. Once a person approves them, tests run against the real application and can be converted into automation scripts with data sets, visual checks and accessibility checks attached. A self-healing function updates element locators when the interface changes. Suites run in parallel from CI systems such as Jenkins, Azure DevOps, CircleCI and GitHub Actions on a lab of over 3,000 real browsers and devices. Supported targets are web, iOS and Android, REST and SOAP APIs, Windows desktop, Salesforce and SAP. Results roll up into coverage and confidence figures.
Key Features
Change Mapping
Links each commit to sprint items, code and requirements so the affected testing surface is visible without manual planning.
AI Test Generation
Writes test cases from commits and sprint context in plain-English steps that can be edited before approval.
Self-Healing
Updates locators automatically when the UI changes, which reduces maintenance of existing scripts.
Cross-Platform Execution
Runs tests on real browsers and mobile devices in the cloud, in parallel.
CI/CD Triggers
Starts runs from the build pipeline and feeds results back to it and to Jira.
Release Confidence Score
Summarises coverage and results as a gate that shows whether a release is ready and which flows are missing.
How Testsigma works
Commit
A code change is mapped automatically to its sprint, stories and specifications.
Generate
Agents write the missing test cases and group them into plans for review.
Execute
Approved tests run against the application and return coverage and confidence results.
Automate and scale
Accepted tests become automation scripts that run in parallel from the CI pipeline.
Release
A gate opens when coverage passes the set threshold, or lists what is still missing.
Screenshots
Use Cases
- Regression Automation
- Move a manual regression suite into scheduled automated runs.
- Sprint Coverage Checks
- See which stories in the current sprint still lack tests.
- Mobile App Testing
- Run the same checks across many iOS and Android devices.
- Salesforce Testing
- Automate tests for Salesforce workflows alongside other applications.
- Release Sign-Off
- Bring a coverage figure and its evidence to the release meeting.
Things to know
- Quote-based pricing
- Both listed plans lead to a custom price request; no rates appear on the page.
- Human approval stays in the loop
- Generated plans are marked pending until someone accepts or rejects them.
- Some features are Enterprise-only
- Accessibility testing, SSO, private grid and on-premise deployment are listed under Enterprise.
FAQs
What is Testsigma?
Testsigma is an agentic test automation platform in which AI agents create, run and maintain automated tests.
What is Testsigma used for?
Testsigma is used by QA and engineering teams to automate functional and regression testing and to judge release readiness.
What types of testing does Testsigma support?
Testsigma supports web, iOS and Android mobile, REST and SOAP API, Windows desktop, Salesforce and SAP testing within one platform.
Does Testsigma require coding?
No. Testsigma lets manual testers and analysts automate without writing scripts, and developers can add custom logic where needed.
Is Testsigma free?
Testsigma offers a free trial, according to its FAQ. Ongoing use requires a paid Pro or Enterprise plan.
How much does Testsigma cost?
Testsigma does not publish fixed prices. Its pricing page offers custom quotes for the Pro and Enterprise plans, based on expected usage such as parallel runs and seats.
Which CI/CD tools does Testsigma integrate with?
Testsigma integrates with Jenkins, Azure DevOps, CircleCI and GitHub Actions, among others, so tests can run on every build.
How does AI work in Testsigma?
Testsigma uses AI agents named Atto for planning, generation, execution, healing and analysis. They draw on connected sources such as Jira, GitHub and Claude Code.
Who it's for
- QA engineers
- Manual testers
- Engineering managers
- DevOps teams
- Salesforce teams
Comments
No comments yet. If you have used Testsigma, tell other readers how it went.
Alternatives to Testsigma
| Tool | Free plan | Starting price | API |
|---|---|---|---|
 TestMu AISoftware testing cloud, formerly LambdaTest, with AI agents that plan, write and run tests for web, mobile and AI apps. |
Yes | — | Yes |
 BrowserStackCloud testing platform for websites and mobile apps on real browsers and devices, with AI agents across the test cycle. |
— | — | — |
 Claude CodeAnthropic's AI coding agent that plans, edits code, runs tests and opens pull requests from your terminal or IDE. |
— | — | — |
 CodeRabbitAI code review that comments on every pull request, with PR triage, change explanations and security monitoring. |
Yes | — | Yes |
 CursorAI coding agent and editor that plans, writes and reviews code on the desktop, in the terminal and in the cloud. |
Yes | — | — |
Listed in the same categories as Testsigma. A starting price appears only where we checked it on the maker’s pricing page; a dash means not checked or not published. All AI Code Assistant tools