MCP server that scans web pages for WCAG accessibility issues with axe-core, so AI agents can audit and fix accessibility. By
Free
Nothing to run from here
Playwright Debug is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
Open Installation—
p95 latency
—
call success
local
runs on your machine
—
calls last 7d
What it does
MCP server for frontend debugging with Playwright - Debug web apps with AI assistance
Quickstart
# 1 — run it from where its publisher ships it
npx -y playwright-debug-mcp
# 2 — ask your agent something
> MCP server for frontend debugging with Playwright - Debug web apps with AI assistance
Playwright Debug is free: there is no plan to choose, no cap to set and nothing that can bill you.
Publishernot claimed
LAAll 2 servers
lars-albinsson2 servers
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Playwright Debug as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. It is scanned, signed and pinned to the version you choose — read the surface before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. Claude Code registers it in one command.
claude mcp add playwright-debug-mcp -- npx -y playwright-debug-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free to install, and it runs from its publisher’s own package: no account, no key, no card.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
62 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Click on an element matching the given selector
Takes no parameters.
Example callreturns →
Fill a form field with the given value. Clears existing content first.
Takes no parameters.
Example callreturns →
Select option(s) from a dropdown/select element
Takes no parameters.
Example callreturns →
Scroll the page or scroll an element into view
Takes no parameters.
Example callreturns →
Wait for an element to appear/disappear or for navigation to a URL
Takes no parameters.
Example callreturns →
Hover over an element
Takes no parameters.
Example callreturns →
Go back to the previous page in browser history
Takes no parameters.
Example callreturns →
Go forward to the next page in browser history
Takes no parameters.
Example callreturns →
Reload the current page
Takes no parameters.
Example callreturns →
Get console log entries from the page. Useful for seeing errors, warnings, and debug output.
Takes no parameters.
Example callreturns →
Clear captured console logs
Takes no parameters.
Example callreturns →
Get network request log. See API calls, their status codes, and timing. Essential for debugging Lambda/API issues.
Takes no parameters.
Example callreturns →
Get detailed info about a specific network request including headers
Takes no parameters.
Example callreturns →
Clear captured network logs
Takes no parameters.
Example callreturns →
Get a summarized DOM tree. Returns tag names, IDs, classes, and text content in a digestible format.
Takes no parameters.
Example callreturns →
Get detailed information about a specific element including computed styles and bounding box
Takes no parameters.
Example callreturns →
Get visible text content of elements matching a selector
Takes no parameters.
Example callreturns →
Take a screenshot of the page or a specific element. Returns base64 PNG image.
Takes no parameters.
Example callreturns →
Execute arbitrary JavaScript in the page context. Powerful for custom inspection and debugging.
Takes no parameters.
Example callreturns →
Get the React component tree with props and state. Works with React 16+.
Takes no parameters.
Example callreturns →
Find React components by name. Returns matching components with their props and state.
Takes no parameters.
Example callreturns →
Get detailed state and props for a specific React component
Takes no parameters.
Example callreturns →
Collect all instances of a React component type with props→HTML mapping. Useful for understanding component behavior, comparing different usages, or extracting component patterns. Returns props, rendered HTML, and minified source for each instance.
Takes no parameters.
Example callreturns →
Inspect Redux store state and recent actions. Works with Redux DevTools or globally exposed stores.
Takes no parameters.
Example callreturns →
Inspect Zustand store state. Finds stores used in React components.
Takes no parameters.
Example callreturns →
Inspect React Query/TanStack Query cache. Shows queries, their status, and cached data.
Takes no parameters.
Example callreturns →
List all React Context providers and their current values.
Takes no parameters.
Example callreturns →
Track component render counts to identify unnecessary re-renders.
Takes no parameters.
Example callreturns →
Find React components with slow render times. Requires React Profiler or DevTools.
Takes no parameters.
Example callreturns →
Get Core Web Vitals (LCP, FID, CLS) and other performance metrics.
Takes no parameters.
Example callreturns →
Detect SSR hydration mismatches. Essential for Next.js/Remix debugging.
Takes no parameters.
Example callreturns →
Find React error boundaries and any errors they have caught.
Takes no parameters.
Example callreturns →
Detect potential memory leaks: large DOM, detached nodes, growing collections.
Takes no parameters.
Example callreturns →
Run accessibility audit: missing alt text, labels, heading structure, color contrast, ARIA issues.
Takes no parameters.
Example callreturns →
Get the accessibility tree showing roles, names, and states. Useful for screen reader debugging.
Takes no parameters.
Example callreturns →
Analyze a console error with full context: related network requests, React state, and debugging suggestions. Best tool for diagnosing issues.
Takes no parameters.
Example callreturns →
Get a quick overview of the debugging state: page info, error count, failed requests
Takes no parameters.
Example callreturns →
Get a token-efficient accessibility snapshot of the page (YAML). Best first tool for understanding page structure — much cheaper than screenshots or DOM dumps.
Takes no parameters.
Example callreturns →
Save the current page as a PDF file (requires headless mode).
Takes no parameters.
Example callreturns →
List all open browser tabs with their URLs and titles.
Takes no parameters.
Example callreturns →
Open a new browser tab, optionally navigating to a URL. The new tab becomes active.
Takes no parameters.
Example callreturns →
Switch the active tab. Subsequent tools target the selected tab.
Takes no parameters.
Example callreturns →
Close a tab (default: the active tab).
Takes no parameters.
Example callreturns →
Configure how alert/confirm/prompt dialogs are answered (default: dismiss). Set before triggering the dialog.
Takes no parameters.
Example callreturns →
Get the log of dialogs (alert/confirm/prompt) that appeared and how they were handled.
Takes no parameters.
Example callreturns →
Set files on a file input element (<input type=
Takes no parameters.
Example callreturns →
List files downloaded during this session with their saved paths.
Takes no parameters.
Example callreturns →
Press a keyboard key (e.g. Enter, Escape, Tab, ArrowDown, Control+A), optionally focusing an element first.
Takes no parameters.
Example callreturns →
Drag one element onto another (drag and drop).
Takes no parameters.
Example callreturns →
Get browser cookies for the current context.
Takes no parameters.
Example callreturns →
Set a browser cookie. Useful for injecting auth/session tokens.
Takes no parameters.
Example callreturns →
Clear all cookies in the browser context.
Takes no parameters.
Example callreturns →
Get localStorage or sessionStorage contents of the current page.
Takes no parameters.
Example callreturns →
Set a localStorage or sessionStorage item.
Takes no parameters.
Example callreturns →
Clear localStorage and/or sessionStorage.
Takes no parameters.
Example callreturns →
Resize the browser viewport.
Takes no parameters.
Example callreturns →
Emulate a mobile/tablet device (viewport, user agent, touch). Recreates the browser context — cookies, storage, tabs, and mocks are reset; the current URL is reloaded.
Takes no parameters.
Example callreturns →
Set the browser geolocation (grants the geolocation permission).
Takes no parameters.
Example callreturns →
Toggle offline mode to test offline behavior and error handling.
Takes no parameters.
Example callreturns →
Throttle network speed to test slow connections (slow-3g, fast-3g, 4g, none).
Takes no parameters.
Example callreturns →
Throttle CPU to test performance on slow devices (rate 1 = none, 4 = 4x slowdown).
Takes no parameters.
Example callreturns →
Navigate to a URL in the browser. Launches browser if not already running.
Takes no parameters.
Example callreturns →
Freeyour plan
$0 forever
- Every tool, no call limit
- No card, no account needed
- Source published under a licence you can read
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because nothing bills
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable. Playwright Debug costs nothing to install and nothing to call, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: $8.42
$16.58 left
ATrust score 93 / 100
Release 0.2.0 · scanned 26 Aug 2026
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
Threat score100
Client score93
Findings1 threat · 1 capability
Capabilityhigh
Coveragesource
Verificationrepo
Surface digest598bc56c6596b7eb9db87329fe7ea31b355c63cc2e480d4cf9a5ffbae51c5519
Score vectorthreat 100 → client 93
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | repo | publisher verification (public source) — no provenance, but the source is public and inspectable | −1 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tool surfaceread
Implementation sourceread
Package metadataread
Live transportnot available
Tools were statically extracted from the published source (39 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| navigate | no tags | |
| click | no tags | |
| fill | no tags | |
| select | no tags | |
| scroll | no tags | |
| wait_for | no tags | |
| hover | no tags | |
| type | no tags | |
| go_back | no tags | |
| go_forward | no tags | |
| reload | no tags | |
| get_console | no tags | |
| clear_console | no tags | |
| get_network | no tags | |
| get_request_detail | no tags | |
| clear_network | no tags | |
| get_dom | no tags | |
| get_element | no tags | |
| get_text_content | no tags | |
| screenshot | no tags | |
| evaluate_js | no tags | |
| get_react_tree | no tags | |
| find_component | no tags | |
| get_component_state | no tags | |
| collect_component_instances | no tags | |
| get_redux_state | no tags | |
| get_zustand_stores | no tags | |
| get_react_query_cache | no tags | |
| get_context_values | no tags | |
| get_render_count | no tags | |
| get_slow_components | no tags | |
| get_web_vitals | no tags | |
| find_hydration_errors | no tags | |
| get_error_boundaries | no tags | |
| detect_memory_leaks | no tags | |
| audit_accessibility | no tags | |
| get_aria_tree | no tags | |
| explain_error | no tags | |
| get_debug_summary | no tags |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
Integrityfirst-seen
Surface digest598bc56c6596b7eb
Scanned2026-08-26
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Pinned to 0.2.0 — the install command below asks for that release. A pin is part of an install, so it is kept for this visit and written down when you install.
0.2.0
undated
installedNo release note was published with this version.
—
no reviews yet
Reviews are earned
Only accounts with at least 50 real tool calls against this server in the last 90 days can post. Ratings are weighted by how much the reviewer actually uses it, and publishers can reply once per review.
Reviews are earned
Writing one takes an account with at least 50 real tool calls against Playwright Debug in the last 90 days. That is the whole gate — there is no other way to post, which is why the counts beside each review are worth reading.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes an account with 50 real tool calls against it.