Dokimore vs testRigor: Plain English or a Recorded Flow?
testRigor removes locators by describing the screen in English. The question is where the test runs, and where the AI does.
Spot something outdated or wrong about testRigor? Tell us — we reply within 5 business days and note every correction on this page.
testRigor is a plain-English test automation platform: its tests are free-flowing English with no locators, the vendor states, covering web, mobile web, native iOS and Android, native Windows desktop, mainframe, API calls, e-mail, SMS and two-factor flows in one test language, with generative AI that drafts tests from a description or from manual test cases, a Chrome recorder, a CLI and an MCP server. It states SOC 2 Type II, ISO/IEC 27001:2022 and HIPAA paperwork for procurement; Dokimore has not obtained any of them.
The comparison is about where things run and what leaves with you. testRigor executes on the vendor’s servers and processes its AI externally, and its exports are its own English language; Dokimore records a flow in a real browser, keeps test intelligence on local models inside your tenant, shows every rescued locator in the run, and exports Playwright, Selenium, Robot Framework or pytest code.
Choose testRigor if…
- Your testers write in English rather than code, and one plain-English language across web, mobile, Windows desktop, API, e-mail and SMS flows is the point. (testrigor.com, checked )
- You want generative AI to draft tests from a feature description or from manual test cases in TestRail, Zephyr, Xray or Azure DevOps. (testrigor.com, checked )
- Vendor-hosted execution and externally processed AI are acceptable, or you will negotiate the on-premise option the vendor offers on request. (testrigor.com, checked )
- Billing by purchased parallel capacity with unlimited users and executions fits your team better than seats. (testrigor.com, checked )
Choose Dokimore if…
- Test intelligence must stay inside your tenant: Dokimore’s models run locally, and no code, prompt or test data goes to an external AI provider unless your team buys and switches on Work with AI.
- You want the exit path in open code: every recording exports as Playwright, Selenium, Robot Framework or pytest, where testRigor exports its own language.
- You need load, OWASP security and WCAG 2.2 evidence from the same recorded flow, with the request and response attached to every confirmed security finding.
- You want every rescued locator in plain sight: the run detail shows the original and the replacement side by side for each healed step.
Capability by capability
- built in
- in the vendor’s core product
- via add-on
- a plugin, an integration or a separately priced product
- not offered
- the vendor’s own pages place it outside the product
- not confirmed
- not described on the vendor pages we read, and not ruled out by them
| Capability | testRigor | Dokimore | Checked |
|---|---|---|---|
| Test authoring | built in Plain English with no locators; generative AI from a description or manual test cases; a recorder delivered as a Chrome extension; a CLI and an MCP server | Record a flow in a real browser (Chromium, Chrome, Firefox, WebKit, Edge) or write code; no plain-English authoring | testrigor.com |
| Self-healing | built in Locator-free tests plus synonyms and opt-in AI healing that labels self-healed test cases fixed-by-ai and puts a warning on each healed step | Fallback locators rescue a step during replay and the run detail shows each rescue; no customer review queue yet | testrigor.com |
| Where the AI runs | built in The vendor states AI data is processed externally through API-based interactions and that it does not currently offer self-hosting or full isolation of AI interactions | Local open-source models inside the platform; no external AI provider unless your team buys and switches on Work with AI | testrigor.com |
| Where tests run | built in Vendor-hosted (its FAQ: it handles everything on its servers); on-premise offered on request | Cloud tenants in India, or a server reserved for your company; no on-premise installer yet | testrigor.com |
| API testing | built in HTTP calls inside end-to-end tests: methods, headers, JSON-path extraction, schema validation, mocking; GraphQL, gRPC, WebSocket and SOAP are not named in the docs | A studio for REST, GraphQL, gRPC and SOAP with environments, chaining, a mock server and microsecond timing | testrigor.com |
| Load testing | not confirmed Listed as a feature using the same end-to-end tests; the docs we read carry no load command or description of the metrics | Profiles from a recorded flow or plain words, P50 to P99.9 and Apdex live; a k6 script export for a profile set up by hand on the load screen | testrigor.com |
| Security scanning | not confirmed Not found on the features page, the language docs or the support site; the vendor’s security pages describe its own programme | OWASP Top 10 (2025) and API Security Top 10 scans, ownership proven first, evidence on every confirmed finding | testrigor.com |
| Accessibility | built in Automated axe checks on every page of a run from one setting; no WCAG version or conformance level stated | axe-core audits against WCAG 2.0 to 2.2 at A and AA, manual criteria never auto-passed, a VPAT draft from the scan | testrigor.com |
| Exit path | built in Tests export as YAML or text in testRigor’s own language; no export to Selenium or Playwright code is described | Playwright, Selenium, Robot Framework or pytest code from any recording; a load profile set up on the load screen as a k6 script | testrigor.com |
| Stated assurance | States SOC 2 Type II, ISO/IEC 27001:2022 and a HIPAA examination; reports on request | No SOC 2 report or certificate; an audit log of administrative changes that owners read through the API; no HIPAA claim | trust.testrigor.com |
| Billing model | Purchased parallel capacity, not seats or runs; a free Public plan where every test and result is public; Enterprise by quote | Flat tiers: Free, Solo Pro $39 a month, Team $249 a month for up to 16 seats, Enterprise from $799 | testrigor.com |
Common questions
Is Dokimore a testRigor alternative?
For teams that want the harness without the English, yes. Both let people who do not write code create tests: testRigor with plain-English sentences on its own cloud and externally processed AI, Dokimore with a recorder in a real browser, local models and every rescued locator shown in the run. If plain-English authoring across mobile, desktop and mainframe is the requirement, testRigor is the better fit; if the exit path must be open code and the AI must stay in your tenant, Dokimore is.
Does testRigor export to Selenium or Playwright?
Not on the pages we read: its exports are YAML or text files in its own plain-English language, and tests run on its servers unless you arrange its on-premise option. Dokimore exports every recording as Playwright, Selenium, Robot Framework or pytest code.
Which one runs tests on my own infrastructure?
testRigor offers an on-premise option on request, while stating that its AI interactions are not self-hostable today. Dokimore has no on-premise installer yet; a server reserved for your company, run by us in India, is the nearest option. Ask each vendor what an on-premise package covers before you compare prices.
Which one has security scanning?
Dokimore. We did not find a customer-facing security scanner on testRigor’s features page, language docs or support site; its DevSecOps pages describe the vendor’s own programme. Dokimore scans the recorded flow for OWASP Top 10 and API Security Top 10 classes with the proving request attached to each confirmed finding.
Go deeper
Other comparisons
Sources
Each page a statement about testRigor was read from, in the order this page cites them, with the date it was read.
- https://testrigor.com/ checked
- https://testrigor.com/how-to-articles/azure-devops-integration/ checked
- https://testrigor.com/faq/ checked
- https://testrigor.com/ai-based-self-healing/ checked
- https://testrigor.com/ai-at-testrigor/ checked
- https://testrigor.com/docs/language/ checked
- https://testrigor.com/features/ checked
- https://testrigor.com/vulnerability-management/ checked
- https://testrigor.com/accessibility-testing/ checked
- https://testrigor.com/test-automation-blueprint/ checked
- https://trust.testrigor.com/ checked
- https://testrigor.com/sign-up/ checked
What changed on this page
- Updated 2026-11-10: Our own cells updated after a change in the product: a load profile now exports from the load screen as a JSON file and as a k6 script, with its test data as a CSV file. The cells and answers on this page that said load profiles have no export are corrected; there is still no JMX import or export. Nothing about testRigor changed.
- Updated 2026-11-10: Corrected our own side. The page said no code, prompt or test data goes to an external AI provider. Since 1 November 2026 a team that buys the Work with AI add-on and switches it on sends redacted text to Anthropic’s Claude models, so the sentence now carries that exception. No statement about testRigor changed.
- Updated 2026-10-25: Re-read against testRigor’s own pages. Its recorder is described as a Chrome extension (we said "Chrome-only", our inference); its AI healing labels self-healed test cases and warns on each healed step; and its tests run on its servers unless you arrange its on-premise option (we said execution "stays on its servers"). Our own description no longer says "reviewed heals", which customers do not have. Every testRigor cell now links the full page it was read from; the trust-centre row keeps its 2026-10-20 date because that site refused our reader this time.