Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
One API key, dozens of capabilities for your AI agent. Zero provider auth.
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
SUPRSONIC_API_KEY, AuthorizationValues marked secret are keys to your account. Never share them or paste them into a chat.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
This address needs a key in a request header, which the add-connector menu cannot set. Use Claude Code, Cursor or VS Code instead.
Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json
{
"mcpServers": {
"suprsonic": {
"command": "npx",
"args": [
"-y",
"suprsonic-mcp"
],
"env": {
"SUPRSONIC_API_KEY": "your-value"
}
}
}
}Run this in your terminal
claude mcp add --transport http suprsonic https://suprsonic.ai/v1/mcp --header 'Authorization: Bearer {suprsonic_api_key}'claude mcp add suprsonic --env SUPRSONIC_API_KEY=your-value -- npx -y suprsonic-mcpFill in the empty values before running it.
Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"suprsonic": {
"url": "https://suprsonic.ai/v1/mcp",
"headers": {
"Authorization": "Bearer {suprsonic_api_key}"
}
}
}
}Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"suprsonic": {
"command": "npx",
"args": [
"-y",
"suprsonic-mcp"
],
"env": {
"SUPRSONIC_API_KEY": "your-value"
}
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"suprsonic": {
"type": "http",
"url": "https://suprsonic.ai/v1/mcp",
"headers": {
"Authorization": "Bearer {suprsonic_api_key}"
}
}
}
}Paste into .vscode/mcp.json in your project
{
"servers": {
"suprsonic": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"suprsonic-mcp"
],
"env": {
"SUPRSONIC_API_KEY": "your-value"
}
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
This address needs a key in a request header, which the add-connector menu cannot set. Use Claude Code, Cursor or VS Code instead.
Your choice stays on this device in a cookie; no account needed.
Authorization secret required HTTP headerThese settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Tool names and descriptions exactly as the server reported them on Nov 7, 2026. The maker wrote them, not us; we show them as plain text.
searchSearch the web with SERP, AI synthesis, or both. Returns, by mode: serp the organic results (title, url, snippet, position) with any knowledge graph and featured snippet; ai a written answer with the URLs it cites; deep both. Use for current facts, news and finding pages. To read
scrapeExtract content from any URL as Markdown or HTML. Returns the page's content in the chosen format (Markdown by default) with its title, description and links. Use when you have a URL and need what the page says. Start with fast; use standard for pages that need JavaScript and tho
profilesFind and enrich professional profiles. Returns the person's name, headline, current company and title, location, summary, experience, education, skills, photo URL and LinkedIn URL. Use to learn who a professional is and what they have done. For their work email use emails; for a
emailsFind professional email addresses. Returns the address with its confidence (0 to 1) and verification status. Use to find a named person's work email at a company. To check an address you already have use email-verify; for the person's background use profiles. Cost: 2 credits, cha
imagesGenerate images from text prompts. Returns a hosted URL of the image and the same image as base64 (b64_json), with its content_type (image/png or image/jpeg, as the bytes are), the aspect ratio and the size used. Use to create a new picture from a description. To cut the backgrou
ttsConvert text to speech audio. Returns the speech as base64 MP3 audio with its sample rate and bit rate (read from the audio itself) and the voice used. Use for spoken voice or narration, up to 2,000 characters a call. For sound effects or music use sound-generate. Cost: 2 credits
sttTranscribe audio to text with timestamps. Returns the transcript, its confidence, language and duration, and every word with its start and end time. Use for a short clip whose exact words and timings you need. For a long recording with several speakers use transcribe; for caption
sound-generateGenerate sound effects and short music tracks from text prompts (NOT speech: use tts for spoken voice). Returns the audio as base64 with its format, sample rate and duration. Use for sound effects, ambience and short instrumental music. It produces no spoken words: for speech use
smsSend SMS or WhatsApp messages. Default channel is SMS (reliable delivery). WhatsApp requires recipient opt-in: they must have messaged your Business number within 24 hours. Returns the message id, channel, recipient, sender number and send status. Sends a real message to a phone
documentsExtract structured data from URLs or content. Returns the extracted data as JSON (shaped by your schema when you pass one), with the source URL and whether long content was truncated. Use to turn a page or a text into structured fields. For a page's full text use scrape; for invo
companiesEnrich company data by domain. Returns the company's name, description, industry, type, size, founding year, headquarters, employees, estimated revenue range, LinkedIn page, social links, logo and brand colors. Use for business facts about a company. For technical facts about its
email-verifyCheck whether an email address takes mail, or is disposable. Returns the status (valid, invalid, unknown when the mailbox is neither confirmed nor rejected, or do_not_mail for a throwaway provider) with a sub-status, whether it is a free mailbox, and the domain's mail servers and
transcribeTranscribe audio with speaker diarization and timestamps. Returns the full transcript with its confidence, language, duration, word and speaker counts, and the utterances, each with its speaker and start and end time. Use for meetings, interviews and podcasts, where who spoke whe
invoice-parseExtract structured data from invoices and receipts. Returns the invoice number, dates, supplier and customer, currency, subtotal, tax, total and every line item. Use for an invoice or receipt given as a PDF or image URL. For any other document use documents with your own schema.
subtitleGenerate SRT/VTT subtitles from audio or video. Returns the subtitles as SRT or VTT text with the language, duration and number of cues. Use to caption a video or an audio file. For a plain transcript use stt or transcribe. Cost: 2 credits, charged only when a call succeeds.
file-convertConvert documents, web pages, spreadsheets and images between formats. Returns a download URL of the converted file with its name, size and formats. Use to change a file's format, for example Word to PDF or PDF to Markdown. To extract specific fields use documents; to read a web
bg-removeRemove the background from any image, at the image's own size or a smaller one you choose. Returns a transparent PNG. Returns the cut-out as base64 and as a hosted URL, with its format, width and height: a PNG with a transparent background, or WebP (transparent) or JPG (set on wh
screenshotCapture a rendered screenshot of any webpage. Returns the image as base64 with its format, its own width and height (a full-page capture is as tall as the page) and its size in bytes. Use to see how a page looks. To read its text use scrape. Cost: 1 credit, charged only when a ca
code-executeExecute code in a secure sandbox with pre-configured environments. Returns stdout, stderr, the exit code, the run time, and the files the code wrote to /home/user/output/ as base64. Use for calculations, data processing and running scripts. Each call runs in a fresh sandbox that
site-intelGet comprehensive domain intelligence: WHOIS, DNS, SSL, tech stack, email security, hosting. Returns, by mode, the registration (WHOIS) data, DNS records with email security, the technology stack, a security grade from A to F, or the subdomains; overview combines registration, DN
researchMulti-step deep research: search, scrape, enrich entities, and synthesize into a cited report. Returns a report (title, summary, sections, key findings, confidence) with numbered sources and the people and companies it found, enriched. Use for a question that needs many sources a
video-infoExtract video metadata, available formats, and thumbnail from any URL. Returns the title, duration, thumbnail, uploader, upload date, view count, platform and the available formats with their sizes. Use to inspect a video before downloading it. To get the file use video-download;
video-downloadDownload video from any URL and return a temporary download link. Returns a download link that expires after one hour, with the file name, size, duration, quality, format, title and thumbnail. A download that takes longer than about a minute and a half (a long video from any site
domainsFind domain names and the extensions that fit a business. Returns, by mode, name ideas with prices, extensions that fit with reasons, one name across extensions, availability for a list of domains, prices per extension, or the extension catalog. Use to name a business or check do
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 8, 2026
The registry name was verified with the GitHub account O-mega-Enterprise, which owns the repo
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
Responded in 677 ms (HTTP 200)
No sign-in needed
Read 24 tool descriptions and found no hidden instructions to the AI
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 3 of 3 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
You need to fill in: SUPRSONIC_API_KEY, Authorization
Either way works: install it on your computer or connect to the cloud version.
The GitHub repo last changed –, and the latest version (0.3.4) reached the registry 20 days ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. It connects by URL, so apps that accept remote MCP servers work too, such as ChatGPT in developer mode.
In the same category: Spec Driven Development, Remote Desktop Commander, Tethered Trust
Servers in the same category from other makers, sorted by stars.
Reading a card:↓ Downloads per weeki★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Fit, compatibility and recall checks for AI agents before they buy: verdicts, evidence, receipts.
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
MCP server for Suprsonic. Gives any AI agent dozens of capabilities through one connection.
Connect to the hosted server, nothing to install:
https://suprsonic.ai/v1/mcp (Streamable HTTP)Authorization: Bearer omk_your_key (or x-api-key: omk_your_key, for a client or gateway that keeps Authorization for its own sign-in)Or run the server on your own machine:
SUPRSONIC_API_KEY=omk_your_key npx -y suprsonic-mcp
Get your API key at suprsonic.ai/app/api. Both offer the same tools at the same prices. Listing the tools needs no key; a call without a valid key answers with the API's own error, which says where to get one.
The package is suprsonic-mcp. Earlier copies of this README named @suprsonic/mcp: that npm scope belongs to an unrelated company, so never install it.
claude mcp add --transport http suprsonic https://suprsonic.ai/v1/mcp --header "Authorization: Bearer omk_your_key"
Add to ~/.cursor/mcp.json (or .cursor/mcp.json in a project):
{
"mcpServers": { "suprsonic": { "url": "https://suprsonic.ai/v1/mcp", "headers": { "Authorization": "Bearer omk_your_key" } } } }
Add to .vscode/mcp.json (VS Code asks for the key once and stores it as a secret):
{
"inputs": [ { "type": "promptString", "id": "suprsonic-key", "description": "Suprsonic API key (omk_...)", "password": true } ], "servers": { "suprsonic": { "type": "http", "url": "https://suprsonic.ai/v1/mcp", "headers": { "Authorization": "Bearer ${input:suprsonic-key}" } } } }
Send the beta header anthropic-beta: mcp-client-2025-11-20 and add to your Messages request:
{
"mcp_servers": [ { "type": "url", "url": "https://suprsonic.ai/v1/mcp", "name": "suprsonic", "authorization_token": "omk_your_key" } ], "tools": [ { "type": "mcp_toolset", "mcp_server_name": "suprsonic" } ] }
Add to the request's tools:
{
"type": "mcp", "server_label": "suprsonic", "server_url": "https://suprsonic.ai/v1/mcp", "authorization": "omk_your_key", "require_approval": "never" }
Suprsonic is listed on Smithery as suprsonic. Add it to your Smithery toolbox and paste your key when Smithery asks for it; Smithery passes it to the hosted server as x-api-key.
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": { "suprsonic": { "command": "npx", "args": ["-y", "suprsonic-mcp"], "env": { "SUPRSONIC_API_KEY": "omk_your_key" } } } }
The same entry works in Cursor's mcp.json.
To serve the HTTP transport from your own machine instead of using the hosted server:
SUPRSONIC_API_KEY=omk_your_key npx -y suprsonic-mcp --http --port 3100
Then connect to http://localhost:3100/mcp.
A call costs credits only when it succeeds. Where a range is shown, the mode you pick sets the price.
| Tool | What it does | Credits |
|---|---|---|
| search | Search the web with SERP, AI synthesis, or both | 1-3 |
| scrape | Extract content from any URL as Markdown or HTML | 1-5 |
| profiles | Find and enrich professional profiles | 3 |
| emails | Find professional email addresses | 2 |
| images | Generate images from text prompts | 3 |
| tts | Convert text to speech audio | 2 |
| stt | Transcribe audio to text with timestamps | 2 |
| sound-generate | Generate sound effects and short music tracks from text prompts (NOT speech: use tts for spoken voice) | 4 |
| sms | Send SMS or WhatsApp messages. Default channel is SMS (reliable delivery). WhatsApp requires recipient opt-in: they must have messaged your Business number within 24 hours. | 1 |
| documents | Extract structured data from URLs or content | 3 |
| companies | Enrich company data by domain | 3 |
| email-verify | Check if an email is deliverable, catch-all, or disposable | 1 |
| transcribe | Transcribe audio with speaker diarization and timestamps | 3 |
| invoice-parse | Extract structured data from invoices and receipts | 3 |
| subtitle | Generate SRT/VTT subtitles from audio or video | 2 |
| file-convert | Convert documents, web pages, spreadsheets and images between formats | 2 |
| bg-remove | Remove background from any image. Returns transparent PNG. | 2 |
| screenshot | Capture a rendered screenshot of any webpage | 1 |
| code-execute | Execute code in a secure sandbox with pre-configured environments | 2 |
| site-intel | Get comprehensive domain intelligence: WHOIS, DNS, SSL, tech stack, email security, hosting | 1-2 |
| research | Multi-step deep research: search, scrape, enrich entities, and synthesize into a cited report | 10-35 |
| video-info | Extract video metadata, available formats, and thumbnail from any URL | 1 |
| video-download | Download video from any URL and return a temporary download link | 2-3 |
| domains | Find domain names and the extensions that fit a business | 1-5 |
Every tool returns a unified response object:
{
"success": true, "data": { "results": [ { "title": "OpenAI raises $6.6B", "url": "https://...", "snippet": "..." } ] }, "error": null, "metadata": { "provider_used": "serperdev", "providers_tried": ["serperdev"], "response_time_ms": 1200, "request_id": "req_abc123" }, "credits_used": 1 }
On failure, success is false and error contains the details (see below).
Every error is the same object. A capability that fails returns it as error in the response above, with success set to false and no credits charged. A refused call (an invalid key, too few credits, a rate limit) answers HTTP 401, 402 or 429 with the object under detail:
{
"detail": { "type": "https://api.o-mega.ai/errors/insufficient-credits", "title": "Insufficient Suprsonic API credits", "status": 402, "detail": "This call needs 2 credits and your Suprsonic account has 0 left of its 200 free credits a month, which refill on November 12. This is your Suprsonic credit balance, not a provider's.", "is_retriable": false, "error_category": "billing", "alternative_action": "Pro gives 2,000 credits a month, 10 times the free plan: choose it at https://suprsonic.ai/app/subscription, or wait until November 12." } }
Error categories: transient (retry safe), permanent (bad request), authentication (invalid key), billing (out of credits), content_access (that URL cannot be served: blocked, region-locked or private).
When using MCP, the AI agent receives the whole error object in the tool response, so it can decide whether to retry from is_retriable and retry_after_seconds, and tell you what alternative_action says to do.
Full API reference with all parameters and example responses: suprsonic.ai/apis