# Shiplight AI visibility in Testing & QA

Canonical: https://devtune.ai/verticals/testing-qa/shiplight

[Website](https://shiplight.ai/)

Latest AI responses: Not available
[Tracked on paid plans](https://devtune.ai/pricing): Google AI Mode, Microsoft Bing Copilot Search, Claude with web search, Grok Search
Prompts: Not available
Runs: Not available


## Platforms



Rank: Not available
Joint rank: Not available
Total brands: 12
Measured responses: Not available
Presence percent: Not available
Share of voice percent: Not available
Average position: Not available
Docs presence percent: Not available
Blog presence percent: Not available
Brand mention percent: Not available


## Profile

Overview: Shiplight AI is an agent-native end-to-end testing platform founded in 2025 and based in San Francisco. It installs into coding agents such as Claude Code, Cursor, Codex and GitHub Copilot as an MCP server and set of Skills. This gives the agent a real browser to verify UI changes while it builds. Those verifications become regression tests written as readable, intent-based YAML and committed to the customer's repository. Tests self-heal when the UI changes, and cached steps replay at Playwright speed. Teams can run them locally, on Shiplight-hosted runners or on their own CI. Results appear on pull requests. The company came out of stealth in February 2026 with backing from Pear VC and Embedding VC. It became self-serve in August 2026. Public customers include HeyGen, Warmly, Jobright and Plaud. Pricing includes a free tier, a $60/month Pro plan and contract-priced Enterprise.
Product summary: Shiplight is an AI-native E2E testing platform that connects to coding agents through an MCP server and Skills. The agent verifies UI changes in a real browser and writes and maintains intent-based YAML tests in the repo. Tests self-heal, run through a CLI locally or in CI, and report results to pull requests and the Shiplight web console.


### Key capabilities

- MCP server and Skills that give coding agents a real browser to verify UI changes
- Intent-based YAML E2E tests stored in the customer's git repo
- Self-healing locators with an AI Fixer for failures
- Deterministic cached replay at Playwright speed, with AI fallback when locators go stale
- Local CLI and visual debugger with screenshots and traces
- Shiplight-hosted or bring-your-own CI runners, with results posted to pull requests
- Test authoring from recordings, tickets, specs or test plans
- Bug triage that reports app bugs instead of rewriting tests to pass
- Email inboxes for testing verification and password-reset flows
- Enterprise options: SSO, audit logs, private cloud/VPC, SOC 2



### Target users

- Engineering teams building with AI coding agents
- QA engineers and QA leads
- Fast-shipping startups and AI-native product teams
- Enterprises needing SOC 2, SSO and VPC deployment



### Key use cases

- Verifying UI changes in a real browser while coding with AI agents
- Auto-generating end-to-end regression suites
- Replacing brittle Playwright selector maintenance with self-healing tests
- Running E2E tests in CI on pull requests
- Triaging test failures as app bugs, test issues or infrastructure problems
- Letting non-engineers contribute tests from recordings or specs

Integrations ecosystem: Shiplight works with MCP-compatible coding agents, including Claude Code, Cursor, Codex and GitHub Copilot, through an MCP server and a /shiplight skill. It runs through the shiplightai CLI (npx shiplight test) locally, on Shiplight-hosted CI runners, or on the customer's own CI. It supports CI providers including GitHub Actions and posts results to pull requests. Customers can bring their own Anthropic, OpenAI or Google LLM key. YAML tests can be transpiled to standard Playwright files, and they run alongside existing Playwright .test.ts files. Enterprise SSO and SCIM offboarding are available on request.
Pricing summary: Shiplight publishes three plans. Free is $0 forever, with a $10 one-time signup credit, 5 members and 30-day retention. Pro is $60/month (a yearly option saves 17%), with $60/month of included credit, pay-as-you-go, auto-recharge, audit logs, 25 members, 90-day retention and priority support. Enterprise is contract priced, with volume pricing, SSO, private cloud/VPC, unlimited members and a dedicated CSM. All plans include the core testing features. LLM tokens and compute are billed at standard usage rates, and customers can bring their own LLM key.
Review summary: Not available
Competitive positioning: Shiplight positions itself as an "agent-native" E2E testing platform for teams that build with AI coding agents. It installs into agents such as Claude Code, Cursor, Codex and GitHub Copilot as an MCP server plus Skills. Tests are intent-based YAML files stored in the customer's git repo. They run on the Playwright engine and can be transpiled to standard Playwright files. This differs from code-first frameworks (Playwright, Cypress), where teams write and maintain the tests themselves. It also differs from established low-code and AI testing suites and managed QA services, which are not built around a coding-agent workflow. Shiplight is a 2025-founded startup, so its maturity, ecosystem breadth and independent review footprint are much smaller than those of the incumbents.
Limitations: Shiplight is a young company, founded in 2025 and publicly launched in February 2026, with about 6 employees. No independent review-platform ratings were found. Usage is credit-based, with LLM tokens and compute billed at metered rates beyond the included credit. The Free plan includes only a one-time $10 signup bonus, 30-day artifact retention and 5 members. SSO and private cloud/VPC are Enterprise-only. The product is built around coding-agent workflows, which may fit less well for teams not using AI coding agents.


### Source urls

- https://www.shiplight.ai/
- https://docs.shiplight.ai/getting-started/what-is-shiplight.html
- https://docs.shiplight.ai/local/yaml-tests/index.html
- https://www.shiplight.ai/pricing
- https://docs.shiplight.ai/cloud_v2/billing.html
- https://www.shiplight.ai/customers/heygen
- https://www.shiplight.ai/blog/what-is-shiplight
- https://www.shiplight.ai/blog/shiplight-vs-playwright
- https://www.shiplight.ai/coding-agents
- https://www.shiplight.ai/enterprise
- https://www.linkedin.com/posts/will-zhao-57372339_ai-softwaretesting-devtools-activity-7424519547619004416-caHt
- https://www.linkedin.com/posts/shrredd_every-engineering-team-i-talk-to-has-the-activity-7424520752491749376-H7ls
- https://www.linkedin.com/posts/will-zhao-57372339_shiplight-is-open-to-everyone-today-coding-activity-7498061677507014656-b4Ie
- https://www.caplight.com/company/shiplight
- https://pitchbook.com/profiles/company/1458937-99
- https://www.everydev.ai/developers/shiplight-ai

Research approved: 2026-11-15T17:50:26.634+00:00


### Customer outcomes

| Customer | Summary | Metric |
| --- | --- | --- |
| HeyGen | HeyGen's Head of QA reports that time spent authoring and maintaining Playwright tests fell from about 60% to almost none within a month after moving to Shiplight's intent-based YAML tests. This is a vendor-published case study. | 60% → ~0% of QA time on Playwright upkeep |
| Jobright | Shiplight's blog says Jobright's co-founder and CTO automated more than 80% of core regression flows within the first weeks. The same source says manual pre-release checks are mostly gone. This is vendor-reported. | >80% of core regression flows automated |
| Warmly | Shiplight's blog says Warmly's Head of Engineering reached reliable E2E coverage across critical flows within days, including complex data-driven logic. This is vendor-reported. | Not available |



### Reviews breakdown





### Review themes



#### Praised





#### Criticized






### Company facts

Founded year: 2025
Hq: San Francisco, California, USA


#### Founders

- Will Zhao
- Feng Qian

Employees range: 1-10
Total funding: ~$7M (per Caplight, unconfirmed)
Valuation: Not available
Arr: Not available
Customer count: Not available
Status: Private


Readiness: Not available


## Ranking





## Platform breakdown





## Strengths





## Gaps





## Topic scores





## Prompt results





## Top sources





## Response excerpts





## Competitor excerpts





## Trend

Visibility delta: Not available
Avg position delta: Not available
Citation count delta: Not available
