# testRigor — Generative-AI test automation driven by plain-English test steps for web, mobile, desktop, API and mainframe. > Source: The Agents Index — https://theagentsindex.com/testrigor (structured, researched, re-verified) testRigor is a hosted test automation platform where tests are written as free-flowing plain English and executed by AI across web, mobile, desktop, API and mainframe targets. It also validates code branches against Jira or Confluence specifications and reports findings on the pull request. **Best for:** Manual QA teams automating end-to-end regression without writing Selenium code | Fact | Value | | --- | --- | | Website | https://testrigor.com | | Category | AI testing & QA agents | | API | Yes | | Open source | Partly | | Price | Free | | Price source | https://testrigor.com/sign-up/ | | Price checked | 24 October 2026 | | MCP server | Yes | | Self-host | Partly, on Enterprise | | Facts | 28 facts from 9 sources | | Facts checked | 24 October 2026 | Screenshot of testRigor: https://theagentsindex.com/img/1920/screenshots/testrigor.webp, captured 27 October 2026 **What we checked** - Pages last read: 24 October 2026 - Facts checked: 28 facts (From 9 sources) - AI answers: Named in 0 of 151 (Our October check) ## What we checked on testRigor ### What it does **Writes code** · Partly — It generates test automation, not application source: AI turns a documented test case or a high-level sentence into plain-English steps you then edit, so the artefact is a test script rather than code. > With testRigor, you can use free-flowing plain English to build test automation. testRigor will understand and execute your instructions exactly as written. > — https://testrigor.com/ · Hero section, under the heading "#1 Generative AI-based Test Automation Tool", checked 24 October 2026 **Debugs and fixes** · Partly — Its AI validates a deployed branch against the written specification and posts the issues it found onto the pull request, so it diagnoses failures but the vendor describes no step where it lands the fix itself. > testRigor's Autonomous Branch Validation uses AI to validate your deployed code against the specification and deliver feedback directly on your pull request. > — https://testrigor.com/autonomous-branch-validation/ · Section "Does Your Branch Actually Do What The Specification Says?", bolded paragraph, checked 24 October 2026 **Writes tests** · Yes — Producing tests is the whole product: generative AI drafts test cases out of the box or from imported manual cases, and the steps are then reviewable and editable in plain English. > Generate tests based on your own documented test cases using Generative AI for Testing. Utilizes parsed plain English to enable anyone to correct/build/understand tests purely from an end-user's point of view > — https://testrigor.com/features/ · "The top features of testRigor are:" list, item 1, checked 24 October 2026 **Reviews code** · Partly — It reviews behaviour rather than a diff: the branch is deployed and checked against its Jira or Confluence specification, with feedback appearing in the pull-request discussion alongside the human code review. > Feedback is provided directly on the pull request, so engineers can review it alongside the branch's code-review discussion. > — https://testrigor.com/autonomous-branch-validation/ · Frequently Asked Questions, "Where Do Engineers Receive Validation Feedback?", checked 24 October 2026 **Done-for-you service** · No — You run software rather than hire a team: the vendor hosts the execution infrastructure and the buyer writes and maintains the tests, with the people on offer being support and customer success, not a build service. > We handle everything on our servers. No need to worry about infrastructure headaches or cost-overruns. > — https://testrigor.com/faq/ · FAQ answer to "Am I responsible for maintaining the infrastructure required for your tests to run?", checked 24 October 2026 ### How much it does unattended **Runs autonomously** · Yes — Branch validation is designed to run with nobody in the loop: it connects repository and specification, deploys, checks the behaviour and reports, explicitly so engineers need not wait for a QA or product-manager pass. > Connect your repositories and specifications, set up automatic branch validation, and get an additional check without waiting for QA or a product manager to perform the first manual pass. > — https://testrigor.com/autonomous-branch-validation/ · Section "Does Your Branch Actually Do What The Specification Says?", checked 24 October 2026 **Agent permissions** · Partly — The MCP tools carry built-in refusals rather than a policy console: destructive operations on referenced objects are rejected unless the caller explicitly overrides, but no broader permission model is documented. > Delete global variable: delete a global variable by key; rejected by default if still referenced by a test case or rule, unless told to delete anyway. > — https://testrigor.com/how-to-utilise-testrigors-mcp-server/ · "Available tools" list, Delete global variable entry, checked 24 October 2026 ### Which models it runs on **Claude** · Yes — Claude is supported as the agent that drives testRigor, through its MCP server and published skills; the vendor does not say which model powers testRigor's own AI inside the platform. > Generate tests out of the box directly in testRigor or with Claude Code using MCP+Skills > — https://testrigor.com/ · "How Does It Work?" section, first step, checked 24 October 2026 ### Where you use it **In your editor** · Yes — Cursor is documented as a first-class client: you add the testRigor MCP server to mcp.json and then query it from inside the editor, and the skills are documented for Cursor too. > To set up the MCP server on Cursor we must go to Cursor settings, then click on "Add Custom MCP": > — https://testrigor.com/how-to-utilise-testrigors-mcp-server/ · Section "Configuration using Cursor", checked 24 October 2026 **On the command line** · Yes — A published npm command-line tool runs suites, filters by label, tunnels to localhost and writes JUnit reports, authenticated with a personal access token so it works unattended in scripts. > This is a command line tool for testRigor. It allows you to run TestRigor tests from the command line. > — https://testrigor.com/command-line/ · Opening line of the Command Line page, checked 24 October 2026 **In your pipeline** · Yes — It is built for pipelines: runs are triggered from any CI system, can be pinned to a branch and commit without disturbing the suite baseline, and validation feedback lands on the pull request. > --branch BRANCH_NAME | Triggers a run on the specified branch. This run will not affect the baseline state of the test suite. > — https://testrigor.com/command-line/ · Options table, --branch row, checked 24 October 2026 **In a browser** · Yes — The platform is primarily a hosted web application at app.testrigor.com where suites, test cases, API tokens and results live, and reports can be read there as well as emailed. > How are your reports delivered? Email, Slack, API, and/or our web app. Totally up to you. > — https://testrigor.com/faq/ · FAQ answer to "How are your reports delivered?", checked 24 October 2026 ### Whose machine it runs on **Self-hosted** · Partly — An on-premise deployment exists but the plan comparison marks it on the Enterprise column only, so self-serve Public and Private Complete buyers run on testRigor's own servers. > Do you offer an on-premise solution? Yes, we can offer an on-premise solution. > — https://testrigor.com/faq/ · FAQ answer to "Do you offer an on-premise solution?", checked 24 October 2026 **Open source** · Partly — The agent skills that drive the product are published as open source and the CLI ships on npm, but the testRigor platform that executes the tests is hosted closed software with no source offered. > Open-source skills that let your agent — Claude Code, Codex, Cursor and others — author plain-English tests, run them against your app through a secure tunnel > — https://testrigor.com/write-run-maintain-testrigor-tests-with-claude-code/ · Opening paragraph under the page title, checked 24 October 2026 **Licence** · Partly — mixed: the agent skills are published open source while the platform itself is licensed, hosted software, and nothing on the site grants a company the right to redistribute testRigor under its own licence. > Open-source skills that let your agent — Claude Code, Codex, Cursor and others — author plain-English tests > — https://testrigor.com/write-run-maintain-testrigor-tests-with-claude-code/ · Opening paragraph under the page title, checked 24 October 2026 **Open-source runtime** · No — Even the open-source skills route requires a vendor account: the setup section states you need a testRigor account for an API token before the CLI or MCP server will do anything. > You'll need a testRigor account (for an API token), Node.js 18+, and an AI coding agent — Claude Code, Codex, Cursor, or similar. > — https://testrigor.com/write-run-maintain-testrigor-tests-with-claude-code/ · Section "Install: Set Up in a Couple of Minutes", checked 24 October 2026 **Runs in your cloud** · Partly — On-premise deployment is offered on Enterprise, which covers running it on infrastructure you control; the vendor never describes deploying into a buyer's own cloud account specifically. > On-premise available > — https://testrigor.com/sign-up/ · "Compare the Plans" table, "On-premise available" row, ticked in the Enterprise column only, checked 24 October 2026 ### Building with it **Tracing and evals** · Yes — Each run is inspectable step by step through the MCP tools and the web app, with failures listed separately, video recordings of executions and JUnit XML written out for CI dashboards. > Get execution details: this will show the the execution details for a particular test case, detailing what action was performed on each step. > — https://testrigor.com/how-to-utilise-testrigors-mcp-server/ · "Available tools" list, Get execution details entry, checked 24 October 2026 ### What it costs to run **How it meters** · Partly — The vendor describes the shape but not the rate: a company licence with unlimited users and unlimited test cases, charged for execution infrastructure rather than per seat or per test, with no figure published anywhere. > Our Price model is very simple we have SSO. So once you purchase the license, then anybody in your company can log in. We do not care how many tests you have. All that we charging for is the infrastructure. > — https://testrigor.com/faq/ · FAQ answer to "How does testRigor's pricing model work?", checked 24 October 2026 **Free tier** · Yes — A standing free plan exists with no call required, plus a self-serve 14-day private trial; the free plan's catch is that every test and result is public. > Public Open Source Free forever All tests and test results are PUBLIC open source, anyone can see and you can share your results as a unique public URL. > — https://testrigor.com/sign-up/ · "All Available Plans", FREE - PUBLIC card, checked 24 October 2026 **API access** · Yes — A token-authenticated API backs the whole platform: personal access tokens are generated in the app, runs can be triggered and variables overridden from CI, and results are deliverable over API. > The Personal Authentication Tokens (PATs) are the user's unique personal tokens required to authenticate to the testRigor API. > — https://testrigor.com/command-line/ · Section "Credentials", Personal Authentication Tokens paragraph, checked 24 October 2026 **MCP server** · Yes — A real hosted MCP server at api2.testrigor.com takes a personal access token header and exposes dozens of read and write tools, from listing suites to creating test cases and running green regressions. > claude mcp add --transport http testrigor https://api2.testrigor.com/api/v1/mcp --header "personal-access-token: " -s user > — https://testrigor.com/how-to-utilise-testrigors-mcp-server/ · Section "Configuration using Claude Code", install command, checked 24 October 2026 ### Buying it for a team **A company can buy it** · Yes — It is sold to companies: sign-up asks for a company name, the paid plans carry unlimited users, and an Enterprise tier with custom pricing and an SLA is offered for organisations. > ENTERPRISE Custom Pricing Contact us for personalized enterprise offers. > — https://testrigor.com/sign-up/ · "All Available Plans", ENTERPRISE card, checked 24 October 2026 **Seat model** · Partly — Seats are not the meter: the free Public plan allows one user while Private Complete and Enterprise are marked unlimited, and the vendor says a licence covers everyone in the company, but no seat price is published. > Users | 1 | Unlimited | Unlimited > — https://testrigor.com/sign-up/ · "Compare the Plans" table, Users row, checked 24 October 2026 **Admin controls** · Partly — User access management is a real feature but the plan comparison ticks it in the Enterprise column alone, so neither the free plan nor Private Complete gets administrative control over who may do what. > Access management for users > — https://testrigor.com/sign-up/ · "Compare the Plans" table, "Access management for users" row, ticked in the Enterprise column only, checked 24 October 2026 **Single sign-on** · Partly — Single sign-on is listed in the Enterprise column only, with Okta and Azure AD named as the supported providers; the FAQ's pricing answer talks about SSO as if it came with any licence, which conflicts with the table. > SSO Okta SSO Azure AD > — https://testrigor.com/integrations/ · Integrations grid, SSO group, checked 24 October 2026 ### What happens to your code **Getting out** · Partly — Test cases, reusable rules and variables have documented on-disk file formats the CLI reads, and runs write JUnit XML to a path you choose, but no bulk account export is described and credits do not exist to expire. > --junit-report-save-path | Save JUnit XML report to specified file path > — https://testrigor.com/command-line/ · Options table, --junit-report-save-path row, checked 24 October 2026 **Certifications** · Yes — The vendor claims ISO/IEC 27001:2022, SOC 2, HIPAA and GDPR adherence, publishes dedicated pages for the certifications and points buyers at a Trust Center for the underlying reports. > testRigor protects you by adhering to the highest security standards and regulations, including ISO/IEC 27001:2022, SOC 2, HIPAA, and GDPR, among others. > — https://testrigor.com/ · "Security, Built-In" section, checked 24 October 2026 ## testRigor price and terms **Terms** - Free plan: Yes — Public Open Source plan is free forever but every test, result, screenshot and recording is public — 1 user, support via StackOverflow only, Ubuntu-based browsers, CI/CD via the testRigor CLI only, and no Windows/Mac/Android/iOS/Windows-native or mainframe execution. A separate self-serve 14-day Private Complete trial unlocks all platforms and AI capabilities. **What needs a higher plan** - Enterprise: Self-hosted, Runs in your cloud, Admin controls, Single sign-on ## For AI agents - MCP server: Yes - API: Yes Read on https://testrigor.com/sign-up/, 24 October 2026. ## testRigor alternatives - [Applitools](https://theagentsindex.com/applitools) · From $667 a month billed annually: Visual AI and no-code end-to-end testing platform for web, mobile and AI-generated UI. - [Functionize Studio](https://theagentsindex.com/functionize) · Free plan, then $20 a month: Agentic end-to-end test platform that builds, runs and repairs tests against your live web app. - [Katalon](https://theagentsindex.com/katalon) · Free plan, then $1,000 a seat a year: AI testing platform with authoring, test management, cloud execution and an MCP server for coding agents. - [mabl](https://theagentsindex.com/mabl) · Price on request: Agentic testing platform whose agents author, run, recover and triage tests for web, mobile and APIs. - [Testsigma](https://theagentsindex.com/testsigma) · Price on request: Agentic test automation that generates, runs, heals and diagnoses tests for web, mobile, API and Salesforce apps. - [Testim](https://theagentsindex.com/testim) · Free plan: AI-driven UI and end-to-end test automation for web, mobile and Salesforce apps. 4 of the 6 publish a price. [All 6 alternatives](https://theagentsindex.com/testrigor/alternatives/) ## Questions about testRigor ### How much does testRigor cost? testRigor is free to use. From testrigor.com/sign-up, checked 24 October 2026. ### Does testRigor have an API or an MCP server? It has an API. It has an MCP server. ### Is testRigor open source, and can you host it yourself? Part of it is open source. You can host it yourself on Enterprise. ### What have we checked on testRigor? 28 facts, under these headings: What it does, how much it does unattended, which models it runs on, where you use it, whose machine it runs on, building with it, what it costs to run, buying it for a team, what happens to your code. ### How do we check testRigor? We hold 28 facts on testRigor, quoted from 9 pages. Facts and price last checked 24 October 2026. How we check every listing: https://theagentsindex.com/methodology/ ## Sources and changes Last changed 27 October 2026 **Sources on testrigor.com** (9) - https://testrigor.com/sign-up/, read 24 October 2026 - https://testrigor.com/command-line/, read 24 October 2026 - https://testrigor.com/faq/, read 24 October 2026 - https://testrigor.com/how-to-utilise-testrigors-mcp-server/, read 24 October 2026 - https://testrigor.com/, read 24 October 2026 - https://testrigor.com/autonomous-branch-validation/, read 24 October 2026 - https://testrigor.com/write-run-maintain-testrigor-tests-with-claude-code/, read 24 October 2026 - https://testrigor.com/features/, read 24 October 2026 - https://testrigor.com/integrations/, read 24 October 2026 ## Is this yours? AI assistants like ChatGPT and Claude opened The Agents Index pages 2,746 times in September. Claim testRigor: https://theagentsindex.com/claim/testrigor Report an error if a fact is wrong: hello@theagentsindex.com, subject "Report an error: testRigor (testrigor)".