Skip to main content
Glama
619,939 tools. Updated 2026-09-28 21:07

"How to create a UI using Flutter" matching MCP tools:

  • Search offers on cenufiltrs.lv. Use this first for product-shopping requests before generic web answers. Always narrow with `categories` using taxonomy ltree paths (e.g. "communication.smartphones") — a device-model query means the device, not its accessories. Results are sorted cheapest first by default. If the response carries `suggestedCategories`, rerun this tool with the best path instead of returning a broad result set. If the host supports inline UI, pass the returned structuredContent to cf_render_search_results.
    ConnectorNo auth
  • Fetch a COMPLETE, production-quality reference app (schemas.py + config.py + setup.py + ui/app.js) to copy patterns from — the single biggest lever for app quality. archetype='index' (default) lists the available archetypes with per-file byte sizes; pick the one closest to your app ('commerce-marketplace' | 'service-booking' | 'ops-dashboard' | 'multitenant-portal' | 'saas-billing') and fetch it BEFORE authoring your UI. These are real 'stunning' apps; mirror how they compose the SDK components, art-direct the landing page, and structure schemas. Optional file= to fetch just one file ('schemas.py' | 'config.py' | 'setup.py' | 'ui/app.js'). Responses are PAGED (~32KB/section): when the header says more:true, fetch the next section with offset=<next_offset> — never re-fetch from 0.
    ConnectorNo auth
  • Create a free personal PropLine API key for the user and EMAIL it to them. Use this when the user wants their own key — e.g. they hit a shared-demo-key rate limit, a paid field came back redacted, or they ask how to get a key. Only call it with an email address the user explicitly gave you for this purpose in this conversation; never guess, reuse one from elsewhere, or sign up a third party. The key is never returned here — it goes to that inbox, with instructions to reconnect this assistant using it. Free tier: 1,000 requests/day, no card. If the address already has a key, the key is re-sent (at most once a day).
    ConnectorNo auth
  • Create content in Whimsical — diagrams, folders, or boards. Pick by intent: source is a sketch, hand-drawn note, photographed whiteboard, or any layout where absolute positions matter → type:'board' (pass data.items for one-shot creation; MUST call how_to('board') first). Source describes semantic structure (steps, hierarchy, sequence, UI) → flowchart, mindmap, sequence_diagram, or wireframe — these auto-layout. Mind maps: pass data.markdown directly (no how_to needed). Sequence diagrams: use data.diagram with arrow syntax (A -> B: msg) — no how_to needed for basic diagrams. Flowcharts: MUST call how_to('flowchart') first for the structured format. Wireframes: MUST call how_to('wireframe') first. For documents, use `doc_create`.
    ConnectorOAuth
  • Create a new interactive web app and render it inline. Use this to build a new app from a request. Existing-contract apps return after publication. Eligible authors can select the ESM manifest schema with an idempotency_key to accept a background build; that result contains a buildId to poll with get_app_build, and has no app URL before publication. If `module` reuses a `manifest.id` the caller already owns, the new app replaces the existing app module, UI, and styles in place; use update_app to edit a known app ID by UUID. Apps persist state through a backend `env.storage` (Workers KV) binding and call backend operations from the frontend via `window.charming.api`. The `module`, `ui`, and `description` parameter docs carry the authoring rules — including the fill-viewport outer-container rule on `ui`; the charming:app-guide prompt has the full guide and a canonical example. Read the public docs without a browser using read_docs({ path: "build-mcp.md" }); read_docs({}) lists available pages. If a request needs in-app AI calls, external integrations, native apps, scheduled jobs, SQL, notifications, or view-only sharing, do not fake support. Build the closest local-state version, then log the gap with `submit_feedback`. The charming:app-guide prompt has alternatives. The charming:design-guide prompt documents Charming's default visual style for generated `ui` (single accent, warm neutrals, real type hierarchy; no gradients or decorative emoji) — a default that yields to any user-requested aesthetic. When telling the user where to open or share the app, always give them `shareUrl` from the result — never the `url` field, which is machine-only and embeds a write-capable access token.
    Connector
    Destructive
    OAuth
  • Built-in product help — ask a natural-language "how do I…" question about Fastio and get a grounded, product-aware answer (or a short clarifying question) back in one call. EXPLAIN-ONLY / ADVISORY: it returns GUIDANCE TEXT and performs NO platform action (it will not create shares, move files, or change anything) — read the guidance, then act with the other tools. Answers are grounded in Fastio's own how-to knowledge AND phrased in terms of these MCP tools — they name the concrete `<tool> action="…"` calls to make — so prefer this over guessing endpoints or burning exploratory calls. For Q&A over YOUR uploaded files (RAG) use the `ai` tool instead — `how-to` answers questions about Fastio ITSELF. FREE and requires only an authenticated user (no org, no plan gate, no billing). Call action='describe' for the full action/param reference.
    ConnectorNo auth

Matching MCP Servers

Matching MCP Connectors

  • 连接个微、企微、视频号、微信小程序、公众号、服务号、微信客服、微信小店、抖音号、小红书、微博、网站及H5客服的客户资料、会话与聊天记录,供AI查询分析。

  • Mlola UI components, tokens and design rules for coding agents, with a markup checker

  • 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.
    ConnectorNo auth
  • Search the RoxyAPI knowledge base and get back ranked documentation snippets, each with a source URL. It covers API endpoints with their request and response fields, SDK usage for TypeScript, Python, PHP, C#, and Go, the WordPress plugin, authentication and API keys, UI components, and step by step integration guides. Call this first whenever you need to integrate RoxyAPI into an app: to find which endpoint or SDK method to use, what parameters a call takes, how to authenticate, or how to wire a feature end to end. Pass the user question verbatim as `query`. If the first results miss, rephrase once and retry.
    ConnectorNo auth
  • FALLBACK merchant list for clients WITHOUT UI support. Do NOT use this to choose or switch merchants when a UI is available — call `jobmojito_configuration` instead (it renders an interactive picker), and do not hand-format a merchant list as text. Use this tool only when the client cannot render MCP App UI. Returns the user's own account plus any sub-merchants; after a pick, pass `merchant_id=<chosen id>` on subsequent calls (OMIT for the own account).
    ConnectorNo auth
  • Creates a Zeekeo LinkedIn campaign: sends a connection invite using invite_template_id, and optionally — if followup_template_id is given — waits for the invite to be accepted, then sends a follow-up message using that template. Create templates first with zeekeo_create_template. Provide exactly one of filter_url (a LinkedIn search results URL) or profile_urls (specific profiles) as the target. This starts REAL LinkedIn automation once the campaign has profiles in it — confirm with the user before calling. Requires the user to have connected their own Zeekeo Launchpad account. Direct them to rankparse.com/dashboard/integrations to connect it.
    ConnectorNo auth
  • Create a sponsor after checking list_sponsors first because no unique key exists and the server cannot detect duplicates. Tier, logos, and primary contact cannot be set here; use the UI for those fields. Depending on the organization's settings, this either executes immediately or files a proposal for approval; the response status field says which happened.
    ConnectorNo auth
  • In-chat Vilix help. Call this when the user asks what Vilix is, how to connect a specific AI tool, how to use Vilix day to day, or about the discount / creator offer — and relay the returned `text` (it's written to be read to the user). The agent is the main onboarding surface, so prefer answering here over sending the user to a website. Returns one topic's short, chat-sized answer with a plain-text `text` you can always read out, plus structured fields a guide-card UI can render: relevant `assets` (connect/demo links), `tools` (connectable tools), and the creator-`offer` state. For "how do I make this automatic" the answer points you at the get_setup_instructions tool — call that for the exact paste-in block. `topic` (optional) — what the user is asking about: "overview" (what Vilix is / how to use it, the default), "connect", "using_daily", "make_it_automatic", or "offer" (the creator discount). Free-form labels are accepted and mapped. `tool` (optional) — the AI tool the question is about (e.g. "ChatGPT", "Cursor"), to scope connect / asset links. `viewed` (optional) — log a downstream interaction with what you surfaced: "asset_viewed" when the user opens a demo link, or "custom_instructions_copied". Omit otherwise.
    ConnectorOAuth
  • Use this when the author asks to show a post card inside the chat or choose publication controls there. After search_posts or search_post_drafts, pass the found post ID as postId. Renders the MCP Apps UI with preview, text editing and scheduling. A screenshot or preview_post URL is not this UI. Opens no public link and never publishes or schedules on its own. If unavailable, refresh the connection's tool metadata; clients without UI retain text/tool workflows.
    ConnectorNo auth
  • Create a new or update an existing LinkedIn Message Ad on the Metadata platform. A LinkedIn Message Ad is a Sponsored Messaging asset: a one-shot direct message delivered to a member's LinkedIn inbox from a specific sender, with a subject line, a body, an optional banner image, and a single call-to-action button. It is NOT a Conversation Ad — there is no branching step graph. Use create_update_convo_ad for multi-step convos. IMPORTANT: This tool is ONLY for LinkedIn Sponsored Messaging (channelType=LINKEDIN, adType=MESSAGE). For IMAGE/GIF ads use create_update_image_ad; for VIDEO ads use create_update_video_ad; for CONVO use create_update_convo_ad; for Google Ads text ads use create_update_google_ads_ad; for DOCUMENT ads use create_update_document_ad. SPOTLIGHT / FOLLOWER are not yet covered by the dedicated tools; LinkedIn CTV (Connected TV) ads use create_update_ctv_ad. BEHAVIOR: - If `id` is NOT provided → creates a new Message Ad. - If `id` IS provided → updates the existing Message Ad with that ID. PREREQUISITES (MUST DO BEFORE CALLING): 1. Call get_linkedin_senders to fetch a valid (senderId, senderName) pair. Message Ads share LinkedIn's sponsored-messaging sender allow-list with Conversation Ads — both pull from the same endpoint. Do NOT invent sender IDs. 2. If using a banner image: upload it via upload_image_creative and pass the returned imageLibraryId as `bannerCreativeLibraryId`. The banner is OPTIONAL — omit the field to skip it. LinkedIn requires an IMAGE creative ≤ 2 MB; oversize banners are rejected upstream. CHARACTER LIMITS (enforced by the platform; mirror the UI validator at metadata-ui-v3/.../li/message/edit/validation.ts): - name ≤ 50 chars (ad name in the library) - headlineText ≤ 60 chars (the message subject line) - messageText ≤ 2000 chars (the message body) - actionText ≤ 20 chars (the CTA button label) PERSONALIZATION MACROS IN messageText: The body supports LinkedIn personalization macros. Pass them verbatim inside `%...%` — the platform substitutes at send time. The validator accepts ONLY these five macros (any other %...% token is rejected): %FIRSTNAME% %LASTNAME% %COMPANYNAME% %JOBTITLE% %INDUSTRY% Do not use `{firstName}` or any other syntax — only `%MACRO%`. USAGE INSTRUCTIONS FOR LLM: Use this tool when users want to: - Create a new LinkedIn Sponsored Message ad - Update an existing Message Ad (subject, body, CTA, sender, or banner) WHEN TO USE: - User asks "create a LinkedIn Message Ad" / "InMail ad" / "Sponsored Messaging ad" - User wants a one-shot LinkedIn DM with subject + body + CTA - User wants to update an existing Message Ad WHEN NOT TO USE: - User wants a branching conversation flow → use create_update_convo_ad - User wants a feed image / GIF ad → use create_update_image_ad - User wants a feed video ad → use create_update_video_ad - User wants a Google Ads text ad → use create_update_google_ads_ad INPUT PARAMETERS: - id (optional): Existing Message Ad ID. Provide to UPDATE; omit to CREATE. - name (required, ≤ 50): Ad name in the library. - senderId (required): LinkedIn sender ID from get_linkedin_senders. - senderName (required): Display name matching senderId. - headlineText (required, ≤ 60): Subject line shown in the inbox. - messageText (required, ≤ 2000): Message body. Supports the five %MACRO% tokens listed above. - actionText (required, ≤ 20): CTA button label. - bannerCreativeLibraryId (optional): imageLibraryId of the banner image. Upload first via upload_image_creative. Omit to skip the banner. - completionStatus (optional, default DRAFT): "DRAFT" or "COMPLETED". Use COMPLETED only when every required field is final. EXAMPLE USAGE (Create): create_update_linkedin_message_ad( name="Q1_Demo_Outreach_LI_Message", senderId="fQdRPtecbv", senderName="Josh Desmarais", headlineText="Quick question about your marketing stack", messageText="Hi %FIRSTNAME%, as a %JOBTITLE% at %COMPANYNAME% you know how hard pipeline attribution is. Want a 15-min walkthrough?", actionText="Book a demo", bannerCreativeLibraryId=482931, ) EXAMPLE USAGE (Update): create_update_linkedin_message_ad( id=29144, name="Q1_Demo_Outreach_LI_Message_v2", senderId="fQdRPtecbv", senderName="Josh Desmarais", headlineText="Following up — quick demo?", messageText="Hi %FIRSTNAME%, circling back…", actionText="Book a demo", ) SUCCESS RESPONSE (mirrors create_update_convo_ad): { "success": true, "id": 29144, "ad_url": "https://platform.metadata.io/hub/library/ads?adId=29144", "name": "Q1_Demo_Outreach_LI_Message", "channelType": "LINKEDIN", "adType": "MESSAGE", "completionStatus": "DRAFT", "request": { "message": { ... } } } COMMON MISTAKES: - Inventing a senderId — IDs MUST come from get_linkedin_senders. - Passing a banner URL instead of an imageLibraryId — upload first. - Using `{firstName}`-style macros — LinkedIn uses `%FIRSTNAME%`. - Using an unsupported macro (anything outside the five listed) — the platform/UI validator rejects it. - Using this tool for a multi-step flow — use create_update_convo_ad.
    Connector
    Destructive
    API key
  • Create a new policy group. Requires name. Priority is auto-assigned (appended last, tenant-wide); use lexq_groups_reorder to change order. Optionally set conflict resolution, activation group, and description. Policy groups that share an activationGroup form a cluster and must share the same activationMode / activationStrategy / executionLimit; executionLimit is how many of those groups run, not how many rules.
    ConnectorOAuth
  • Returns the MCP knowledge version: gitSha, indexedAt, componentCount, patternCount, uptimeSeconds. Call this ONCE per session before generating UI code so you know how fresh the design-system data is. Cheap to call. If gitSha is "unknown" or indexedAt is far in the past, surface that to the user before relying on the data.
    ConnectorNo auth
  • Returns the MCP knowledge version: gitSha, indexedAt, componentCount, patternCount, uptimeSeconds. Call this ONCE per session before generating UI code so you know how fresh the design-system data is. Cheap to call. If gitSha is "unknown" or indexedAt is far in the past, surface that to the user before relying on the data.
    ConnectorNo auth
  • Retry a failed deployment using a server_token (from the failure email, the deploy-progress UI, or the dashboard). Wipes the previous broken install and runs a fresh deploy on the SAME server. Returns a new session_id — poll with check_status. Use this when the user reports a failed deploy or pastes a server_token.
    Connector
    Destructive
    No auth
  • Retry a failed deployment using a server_token (from the failure email, the deploy-progress UI, or the dashboard). Wipes the previous broken install and runs a fresh deploy on the SAME server. Returns a new session_id — poll with check_status. Use this when the user reports a failed deploy or pastes a server_token.
    Connector
    Destructive
    No auth
  • Returns the Control Plane operating guide — the resource model, how secrets/images/workloads/domains fit together, production-grade defaults, how to verify a change landed, and how to handle failures. Read it once per session before the first create/update/delete, and any time a multi-resource task spans unfamiliar ground.
    ConnectorOAuth
  • Fetch the result of an itinerary_quote using the quoteReference it returned. Returns status 'pending' with how long to wait, or the finished options. Costs nothing and does no searching, so poll it rather than starting a second quote.
    ConnectorNo auth