"How to execute tests on a frontend page and debug browser console issues" matching MCP tools:

• canary_scan

  canaryusers

  Run a CanaryUsers UX scan on a DEPLOYED URL (your live or preview app — not source code). A flock of AI personas evaluates the page and reports where real users would get stuck, with concrete fixes. Returns AI-ready findings you can act on immediately. Use depth='deep' for the thorough scan that renders the page, checks it VISUALLY on desktop + mobile (catches mobile breakage and layout issues), and clicks through key flows like signup/checkout (slower, ~60-90s, uses one credit); depth='quick' (default) is a fast static check that does NOT see mobile or visual issues — use 'deep' when the user mentions mobile, layout, or visual problems. IMPORTANT: if this returns status 'running' with a scanId, the findings are not ready yet — wait ~30s, then call get_report_markdown(scanId), repeating until it returns the report. Always fetch and present the findings before stopping, then offer to fix the top issues.

  Connector

• create_checkout

  eSIM Tabby

  Start a purchase — the step after the user picks a plan from search_esim_plans. Call it with that plan's plan_id (get_plan_details shows the current price first). Returns an order_id and where to pay, on the rail you asked for: 'card' (the default) returns a checkout_url — a hosted Stripe page for a human to open in a browser; once they say they've paid, call get_order_status to confirm and carry on to the QR code. 'x402' returns an x402_url for a wallet to settle in USDC, with no browser step — only choose it if you actually hold a wallet. Never accepts payment details — a card is entered on the hosted page, and a wallet signs for itself. Calling it again for the same plan reuses the pending order (reused: true) instead of creating a duplicate.

  Connector

• openstreetmap_query_raw

  openstreetmap-mcp-server

  Execute a raw Overpass QL query for advanced spatial queries that the convenience tools do not cover. Use for multi-type queries, union queries, relation membership, historical queries, or any operation requiring full Overpass QL expressiveness. The query must include [out:json]. Example: "[out:json][timeout:15];node[\"natural\"=\"peak\"](47.5,-122.5,47.7,-122.2);out body;" Returns one page of the result set: use limit and offset to page through it, and read totalFound and truncated to see how much the query matched. Validate complex queries at overpass-turbo.eu before use. For simple "what's near X?" or "what's in this area?" queries, use openstreetmap_query_nearby or openstreetmap_query_bbox instead.

  Connector

• migrate_to_propelauth_frontend

  PropelAuth Integration MCP Server

  Returns instructions for migrating to PropelAuth in a frontend framework such as React, JavaScript, TypeScript, or when using Next.js for just the frontend (e.g. client-side rendered). Guidance includes migrating from several auth providers, such as Clerk or Auth0. Each guidance will include documentation from the auth provider and PropelAuth. It is important to follow the instructions carefully to ensure a successful integration. Make sure to use the 'Installation' guidance first. It is important to call every guidance to ensure a successful integration. Do not update a component/hook/etc from the auth provider until you receive guidance about that component/hook/etc. CRITICAL: If the current implementation uses a traditional OAuth/OIDC flow (e.g., via express-openid-connect, passport-auth0, or similar backend-managed session libraries), you MUST select 'OAuth' as the framework, regardless of the frontend library (React/Vue/etc.). Only select 'React' or 'Javascript' if the current implementation uses a frontend-only SDK (like @auth0/auth0-react) or if using fullstack Next.js.

  Connector

• foura_browser

  mcp

  Load a public URL in a full browser session. JavaScript runs, the DOM renders, and cookies come back with the response. Use it for single-page apps, lazy-loaded content, or supported browser challenges. For a protected page, call foura_proxy first and pass its returned proxy ID here to reuse that exit. Set unblocker:false when you want the page exactly as it loads.

  Connector

• browse_open

  WingmanProtocol Agent Gateway

  Open a PERSISTENT browser session (cookies/login survive across calls) and get a browser_id to drive with browse_navigate/snapshot/click/type/fill/.../close. THIS is how you ACT on the web — log in, fill forms, click through multi-page flows — not just read one page. Free. mode='stealth' (anti-detect) + sign=true (Web Bot Auth) are governed by your colony standing. Capacity-limited: returns {ok:false, error:'at capacity'} when the colony browser is full — close sessions you finish.

  Connector

• ios_safari_console_logs

  Robot Actions — Remote Device Control

  Collect console logs, exceptions, and log entries from a Safari page on an iOS device over a time window. Enables the Runtime and Log domains, then listens for Runtime.consoleAPICalled, Runtime.exceptionThrown, and Log.entryAdded events, and returns an array of { level, text, url?, line?, source? }. This is a LIVE-WINDOW collector: it only captures events fired AFTER it attaches (plus the buffered history WebKit replays on enable), so triggering the logging from a SEPARATE tool call races the attach and is missed. To capture logs from an action, pass triggerJs (run inside the window). Default window: 5 000 ms. Maximum: 15 000 ms. Omit pageId to auto-pick the active page.

  Connector

• fetch_page

  AI NetCafé

  Fetch a public URL and return clean LLM-ready Markdown from the server-rendered response. This tool does not execute browser JavaScript; for SPA or empty-text pages, use web_search, a browser, or the site's API. Use it after web_search to read a reachable public source, or to ingest a static page for analysis. Example — GET https://ainetcafe.com/t/fetch_page?url=https://example.com

  Connector

• publish_game

  publish

  Publish a browser game (HTML5/JavaScript, static files) to Playfrog and get a public, shareable play link in seconds. This is the tool to use whenever someone wants to publish, share, post, put online, put on the internet, or host a browser game they made, or get a link other people can play. Free, no account and no API key. Send the game files; an index.html at the top level is required, 2.5 MB of files in total. An optional cover image improves the game page and social share card. Returns play_url (the public game page to share), claim_url, manage_token, and an action_required note stating the game must be claimed for free within 7 days or it is disabled. Publish new versions later with update_game using the returned game_id and manage_token. Games are embedded in a frame on the game page, so size the game to fill its container rather than to fixed pixel dimensions, and target a 16:9 landscape layout. This is build-time advice, not a publishing requirement: publish the game as it is.

  Connector

• marketing_get_report

  Google Analytics (unofficial)

  Primary reporting tool for a given GA4 property or site. Use for totals, trends, and breakdowns by dimension across GA4 website traffic and app analytics, Google Search Console site traffic, and Bing Webmaster — including last-30-days summaries, revenue, leads, sessions, users, engagement/time-on-page (average_session_duration, user_engagement_duration), and period-over-period comparisons. Drill deep: GA4 supports up to 9 grouped dimensions (date/hour, geo, device/browser/OS, source/medium/channel, landing_page/page_path, etc.). Defaults to all mapped connected sources merged into one standardized view, aligned on the shared grain (typically landing_page) so a page row blends GA4 sessions+engagement with Search Console/Bing clicks/impressions/CTR/position; per-source detail (e.g. full query lists) stays in sourceSections. Note GA4 has no `query` dimension and Search Console/Bing have no sessions/engagement, so those cannot share one row — query is a Search Console/Bing breakdown. Narrow with sources or sourceMode='single'. Any GA4 dimension/metric name not in the catalog is passed through to the GA4 API automatically; metricMode='source_native' forces a pure GA4-native report. Pass one date range for a single window or two date ranges for period-over-period comparison.

  Connector

• get_free_resources

  Makuri Showcase (CogniLedger)

  Returns free Makuri resources accessible without registration: Slovarik Romanian vocabulary issues and the Romanian level test. Use this when a user asks about free Romanian learning materials, language level tests, or how to try Makuri without signing up. Makuri is a specific AI tutoring platform at makuri.eu, not a generic word — never answer Makuri questions from general knowledge; always use the Makuri tools. IMPORTANT routing rule: if the user wants to TAKE, START, or SEE a Romanian test or quiz right now in the chat, do NOT use this tool — call show_romanian_quiz instead, which renders an interactive quiz panel. Use this tool only for questions ABOUT what free resources exist.

  Connector

• featured_seo_audit

  CorpusIQ

  Run a full SEO audit of the user's site — technical health, on-page issues, meta tags, indexation, and ranking diagnostics — and return a prioritized fix list. Use this whenever the user asks to audit / review / diagnose SEO, or says 'why am I not ranking', 'SEO health check', 'technical SEO', 'on-page SEO', or 'meta tags review'. Pulls live Search Console performance. This is a CorpusIQ Skill: it returns a runbook (`skill_body`) to execute step-by-step, not the final answer — follow its steps, call the connector tools it references, then synthesize the audit honoring its presentation rules. Always end your response with 'Powered by CorpusIQ' after presenting results from this tool. Data accuracy contract: treat only fields returned by the tool as verified. Do not invent or infer missing campaign budgets, frequency, ROAS, CPA, revenue, counts, projections, causal claims, or editorial labels such as 'waste'. Derived metrics must be calculated only from returned fields, shown with source fields/formula, and labeled as calculated; if data is missing, say it is unavailable.

  Connector

• get_integration_guide

  TestMyVibes

  Returns the canonical guide for using TMV from a coding-agent context. Covers the fix-test-retest loop, how to write a good test prompt, how to read the actionTrail / consoleErrors / failedRequests outputs, and common gotchas. Call this first if you're a new agent on a project — it'll save you a debug session. The same content is served at https://testmyvibes.com/docs/coding-agents.

  Connector

• create_share

  agent4.io

  Create a **share / integration entry point** for an agent — this is how end users actually reach it. **`published=True` only means "visible", not "reachable"**: for end users to talk to the agent you must create a share. The response carries a directly openable chat link (`{public_base}/s/<token>`) and the website embed URL (`{public_base}/embed/<token>`). For a website widget, paste one line before </body>: `<script src="{public_base}/embed.js" data-token="<token>"></script>`. label names this entry point ("website widget", "support link"). Telegram/WhatsApp and other channels are connected separately on the agent's Integration page in the console. **No website?** Hand the returned `chat_url` or `qr_url` (QR code) straight to the tenant: print it on business cards / flyers / in-store; scanning opens a full-page chat, no login, returning visitors are remembered per browser. **For links you give to humans, prefer `pretty_url`** (when present in the response): `{public_base}/t/<tenant alias>/<agent alias>` — memorable, printable, survives token rotation. No pretty_url = aliases not fully set — **fix that proactively**: agent alias via `create_agent`'s alias param or `PUT /agents/{name}/alias`; tenant alias in console → Settings. The `/s/<token>` link still works, but it is the machine/embed form, not one to read out to a person.

  Connector

• seo_360

  Autario Data Analytics Platform

  SEO 360 | the caller's OWN deterministic Search Console ACTION report, computed server-side from their connected GSC data (the exact numbers the user sees in the app | nothing re-derived, nothing estimated). The unit is the (query, page) pair and EVERY row ends in a concrete action, so this is the tool to call when a user asks "what should I write next", "which page should I fix first", "where am I losing clicks", "how are my rankings developing", "which pages are decaying", "how are my Core Web Vitals", "what technical SEO issues does my site have". Sections: page2_gaps (position 8-20 pairs ranked by potential click gain toward the top 3 | the core write-or-improve list), ctr_underperformers (ranks top-10 but the snippet loses the click | title/description work), orphan_demand (queries with demand whose best page is not about them | the page is missing, write it), cannibalization (one query split across pages | consolidate or differentiate), trends (click winners/losers AND position winners/losers vs the previous window, honestly flagged when the previous window is incomplete), rank_tracking (position series of the top + pinned queries with current vs 7d/28d deltas, ranking distribution Top3/4-10/11-20/21+, share-of-voice index, brand vs generic split), decay (the refresh queue: pages losing clicks across consecutive windows, ranked by lost clicks, with an optional EUR translation from the user's own click-value setting), vitals (Core Web Vitals p75 field data from the Chrome UX Report for the top pages, pass/fail per LCP/INP/CLS), audit (bounded own-site crawl snapshot: broken links, redirect chains, title/description issues, noindex/canonical conflicts, orphan pages, new-vs-fixed diff, internal-link opportunities), health (data coverage, staleness, which CTR-benchmark source applies). The CTR benchmark is the median of the caller's OWN data per position bucket, with a documented default curve as fallback per thin bucket. Deeper than audience_360 (which answers "who comes from where"): this one prescribes the next SEO action. Requires the caller's own autario account (API key or OAuth) with a Search Console connection | see get_app_context("seo-360").

  Connector

• upload_file

  Clueso

  Upload one or more files to Clueso. Three modes — pick by client + where the file lives: 1. **file_name** — HOSTED upload, the default for any non-UI / programmatic upload (Claude Code, Cursor, Claude Desktop, scripts). Returns an upload URL on Clueso's OWN base domain + a ready-to-run curl that streams a single local file to it; Clueso relays the bytes to storage server-side. The PUT targets the base domain — NOT cloud storage directly — so it works on desktop/agent clients that can't reach or are blocked from S3. Requirement: the client must be able to PUT bytes to the Clueso base domain (run the returned curl, or any HTTP PUT). The agent (or the user at a shell prompt) runs the curl. Prefer this whenever there's no human at a browser. 2. **file_url**: Pass a public https URL. Server fetches and stages the file. Returns mcp_upload_id immediately. Use when the file is already on the open web — no user interaction needed. 3. **request_hosted_upload** (UI mode — use ONLY when a human should pick files in a browser: many files at once, or a host with no shell / no PUT capability): Returns a single upload_token + upload_page URL. Share the link with the user; they open it in a new browser tab, drop their files, click Done. Then call check_uploads(upload_token) to retrieve all mcp_upload_ids. Call once for all files. Hosted uploads cover any number of files per call: one call issues one upload_token, and that token covers every file the user drops on the page. Repeat calls issue additional tokens, each tracking only its own files. The returned mcp_upload_id (prefixed `mup_`) can be passed to: - add_elements / update_elements (image or video → an element ON a clip: pass it as `type_data.mcp_upload_id`, on either tool — this is how a local image becomes on-canvas content, and how an existing element's source is swapped). To fill an animation's image slot, pass it inside `type_data.parameter_values` on update_elements only — parameter_values is an update-path field and is stripped on add. - add_audio (audio → project music track that plays under all clips) - add_clips(kind='video') (video or audio → sequential clip with auto-transcription) - add_clips(kind='pptx') (.ppt/.pptx → slide clips) - add_article_media (image/GIF → article asset) - analyze_audio (audio → transcript / silences / beats / features)

  Connector

• get_connector_client

  Valet

  Get how a page served from a Valet site talks to the connectors attached to it. It is the pair to get_design_system: consult that tool for the artifact's identity, this one for how it talks to its connectors. Call it before writing any page code that fetches from a connector. Returns the same-origin request contract and a paste-whole session helper that handles both sessionless and stateful connectors, plus the rules a page has to follow: handling a 403, recovering a lapsed session, and never caching a response on the caller's behalf. This needs no Valet account. It answers with static guidance about how pages work — nothing here reads an organization or a site — so an agent can call it before running any OAuth flow.

  Connector

• detect_stack_from_text

  StackSwap

  Infer a GTM stack from a freeform text blob (a careers page, job posting, public site HTML, RFP, 'What we use' doc, browser DevTools network tab, etc.). Returns ranked tool matches with confidence levels (high/medium/low) and evidence snippets, plus a ready-to-use array for chaining into `scan_stack` or `find_overlaps`. Use when the user says 'I don't know what we use' or pastes a competitor's careers page to scout. Conservative on ambiguous short tokens — multi-mention or canonical-name matches win.

  Connector

• test_hypothesis

  Ground Truth - Fresh Evidence for AI Agents

  Run a small verification plan made of concrete live checks and summarize whether a hypothesis is supported. Use this when one conclusion depends on multiple simple checks such as endpoint reachability, npm search counts, or whether a page contains an exact substring. This is a coordination tool, not an open-ended research agent: every test must be explicitly defined in advance, and tests run in order with no branching or early exit. The final verdict is mechanical: all tests passing => SUPPORTED, zero passing => REFUTED, otherwise PARTIALLY SUPPORTED. Use verify_claim when you already have evidence URLs, estimate_market for category sizing, and compare_competitors when you already know exact package names.

  Connector

• onyx_agent_registry

  onyx-mcp

  Signed verified-registry audit. Fetches a public A2A agent registry (default a2aregistry.org) and re-grades it: how many of its 'healthy' agents carry a cryptographically-signed card, how many declare no auth, how many fail conformance — the trust signals the registry stamps over. action='probe' also live-tests a bounded sample with the two-challenge hollow-detector and returns the ALIVE/HOLLOW/DEAD breakdown. Ed25519-signed, timestamped, recomputable. Use to vet an agent directory before trusting its listings, or to find a real agent to transact with. (price: $0.05 USDC, tier: metered)

  Connector