Holmes
By Holmes AI BV. Category: Agentic end-to-end testing. Homepage
According to its documentation, Holmes is an autonomous QA platform that learns how people use a web app, turns those journeys into tests and runs them continuously. A test is a document with a goal and a list of checks; tests can be drafted with Holmes in plain English, discovered automatically by agents exploring the app, or imported from files such as PDF, Word, Excel or CSV. Runs can be manual, scheduled or triggered from CI/CD events. The docs describe native Android app testing on cloud emulators, with iOS not supported, and an MCP server for AI coding assistants such as Claude Code. The homepage lists Freemium, Basic and custom-priced Professional plans.
Description written from the vendor's documentation.
Status
Status: Active
Facts
| Fact | Value | Evidence | Source | Verified on |
|---|---|---|---|---|
| Status | Active | Vendor-stated | useholmes.com | |
| Founded | 2026 | Vendor-stated | useholmes.com | |
| Headquarters | Ghent, Belgium | Vendor-stated | useholmes.com | |
| AI capabilities | Documented | useholmes.com | ||
| Test creation | Documented | useholmes.com | ||
| Test output | Proprietary runner | Documented | useholmes.com | |
| Surfaces | Documented | useholmes.com | ||
| Hosting | Vendor-stated | useholmes.com | ||
| Pricing model | Vendor-stated | useholmes.com | ||
| Free tier | Yes | Vendor-stated | useholmes.com | |
| CI integrations | Documented | useholmes.com | ||
| Self-healing | Vendor stated | Vendor-stated | useholmes.com | |
| Agent integration | Holmes MCP server (OAuth 2.1) for AI coding assistants such as Claude Code; tools include list_products, list_runs, list_issues, get_issue, create_suggested_flows, list_flows and update_flow to read QA state, propose tests, edit tests from plain-English instructions and resolve issues | Documented | useholmes.com |
Published prices
| Plan | Price | Billing |
|---|---|---|
| Freemium (EUR display; 1 product, 5 saved flows, 5 runs / month) | €0 | not stated |
| Freemium (USD display; 1 product, 5 saved flows, 5 runs / month) | $0 | not stated |
| Basic (EUR display; unlimited products, 30 runs / month) | €499 per month | not stated |
| Basic (USD display; unlimited products, 30 runs / month) | $549 per month | not stated |
Vendor-stated, verbatim
What Holmes AI BV says about its product, quoted as published and not confirmed in documentation.
“Set up in under 10 minutes. First results in under an hour.”
“Under 10 minutes. Add your URL, connect a test account, and Holmes starts its first run. Most teams see their first results within an hour.”
“Your team fixes bugs in minutes, not days.”
“What used to take a QA team days now runs in parallel across AI agents. Results arrive before your PR review is done.”
“No brittle selectors, no flaky tests. When your UI changes, Holmes adapts its flows automatically — zero maintenance overhead.”
“Holmes reproduces each issue step by step, with screenshots and video. No noise, no false positives, just problems worth fixing.”
Alternatives
Tools in the same category (Agentic end-to-end testing), ordered by the number of shared creation and surface values, then by name. Each line lists documented facts that differ from Holmes.
- Pie: Agentic end-to-end testing; no documented difference in output, creation, hosting or free tier
- QualityMax: documents output Mixed; creation by code generation; hosting self-hosted, hybrid
- Quorvex AI: documents output Standard Playwright; creation by code generation; hosting self-hosted
- Robonito: documents creation by prompt to test, recording
- ACCELQ: documents creation by recording; hosting self-hosted, hybrid
- Autify: documents output Mixed; creation by prompt to test, recording; hosting self-hosted
Questions
Is this page a review of Holmes?
No. This page is a fact sheet. Each fact shows its evidence level, its source and the date it was checked. We do not rate or rank tools.
Does Holmes have a free tier?
Yes. Holmes has a free tier. This is stated by the vendor and not confirmed in documentation, checked on 2026-09-24.
What test output does Holmes produce?
Output: Proprietary runner. This is documented, checked on 2026-09-24.
Is Holmes still available?
Status: Active, checked on 2026-09-24.
Sources
- Holmes - AI QA Detective (homepage, including pricing and FAQ sections) Opened
- What is Holmes? (Holmes docs) Opened
- Adding a test (Holmes docs) Opened
- Mobile app testing (Holmes docs) Opened
- Use Holmes from Claude Code (MCP) (Holmes docs) Opened
- Webhooks (Holmes docs) Opened
- Holmes press kit Opened
- Holmes terms of service Opened
Spotted an outdated or wrong fact about Holmes? Send the source and we apply it at the next monthly update.