Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
CAPTCHA and anti-bot solving for Turnstile, DataDome, Akamai, PerimeterX and Imperva.
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.
SOLVEHUB_API_KEY, SOLVEHUB_BASE_URLValues 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 …”.
Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json
{
"mcpServers": {
"solvehub": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"SOLVEHUB_API_KEY",
"-e",
"SOLVEHUB_BASE_URL",
"ghcr.io/seencxy/solvehub-mcp:0.1.0"
],
"env": {
"SOLVEHUB_API_KEY": "your-value",
"SOLVEHUB_BASE_URL": "your-value"
}
}
}
}Run this in your terminal
claude mcp add solvehub --env SOLVEHUB_API_KEY=your-value --env SOLVEHUB_BASE_URL=your-value -- docker run -i --rm -e SOLVEHUB_API_KEY -e SOLVEHUB_BASE_URL ghcr.io/seencxy/solvehub-mcp:0.1.0Fill in the empty values before running it.
Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"solvehub": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"SOLVEHUB_API_KEY",
"-e",
"SOLVEHUB_BASE_URL",
"ghcr.io/seencxy/solvehub-mcp:0.1.0"
],
"env": {
"SOLVEHUB_API_KEY": "your-value",
"SOLVEHUB_BASE_URL": "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": {
"solvehub": {
"type": "stdio",
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"SOLVEHUB_API_KEY",
"-e",
"SOLVEHUB_BASE_URL",
"ghcr.io/seencxy/solvehub-mcp:0.1.0"
],
"env": {
"SOLVEHUB_API_KEY": "your-value",
"SOLVEHUB_BASE_URL": "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 one runs on your computer, and ChatGPT can only connect to servers on the internet. Pick another app.
Your choice stays on this device in a cookie; no account needed.
SOLVEHUB_BASE_URL envThese settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Runs on your computer; the tool list appears in your AI app after you install it.
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 9, 2026
The registry name was verified with the GitHub account seencxy, which owns the repo
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
This one runs on your computer; there is no internet endpoint
Runs on your computer and uses local settings instead
Runs on your computer; we can only read tool lists from internet endpoints
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 1 of 1 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: SOLVEHUB_API_KEY
It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).
The GitHub repo last changed –, and the latest version (0.1.0) reached the registry yesterday.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Docker on your computer.
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 the SolveHub CAPTCHA and anti-bot solver API. Gives Claude, Cursor, VS Code and any other MCP client eight tools to solve a challenge and get the token, cookie or OCR text back - Turnstile, Interstitial, DataDome, Akamai, PerimeterX, Imperva, AWS WAF, Alibaba CAPTCHA 2.0, Arkose FunCaptcha and image OCR through a single REST endpoint.
Zero dependencies, one npx command, Node 18 or newer.
| Tool | What it does |
|---|---|
solve_captcha | Create a task and wait for it, returning the solution (token, cookies or text). |
create_task | Create a task and return its id immediately, for work you finish later. |
get_task | Status of one task, plus its solution once done. |
wait_for_task | Poll an existing task until it is done, failed or cancelled. |
cancel_task | Cancel a queued or running task; the reserved amount is refunded. |
list_tasks | Recent tasks, newest first. |
get_balance | Available and reserved balance in USD. |
get_prices | Live per-task price for every challenge type. |
Get an API key from the SolveHub console; new accounts get a $0.10 trial credit after Telegram verification.
Edit claude_desktop_config.json and restart the app:
{
"mcpServers": { "solvehub": { "command": "npx", "args": ["-y", "github:seencxy/solvehub-mcp"], "env": { "SOLVEHUB_API_KEY": "sh_live_xxx" } } } }
Most clients read the same JSON in mcp.json (VS Code and Cursor use the servers key):
{
"servers": { "solvehub": { "type": "stdio", "command": "npx", "args": ["-y", "github:seencxy/solvehub-mcp"], "env": { "SOLVEHUB_API_KEY": "sh_live_xxx" } } } }
git clone https://github.com/seencxy/solvehub-mcp
cd solvehub-mcp SOLVEHUB_API_KEY=sh_live_xxx node index.mjs
A public image is built from this repo on every push to main, so no local Node install is needed:
docker run -i --rm -e SOLVEHUB_API_KEY=sh_live_xxx ghcr.io/seencxy/solvehub-mcp:latest
Claude Desktop (or any stdio client) can launch the same image:
{
"mcpServers": { "solvehub": { "command": "docker", "args": ["run", "-i", "--rm", "-e", "SOLVEHUB_API_KEY=sh_live_xxx", "ghcr.io/seencxy/solvehub-mcp:latest"] } } }
Solve the Cloudflare Turnstile on https://example.com/login. The site key is 0x4AAAAAAA...
and I am allowed to test this site.
The model calls solve_captcha with type: turnstile and the challenge fields, then answers with the token and its expiry. Anything that needs a proxy (DataDome, Imperva, Akamai, PerimeterX) takes a proxy argument as well, for example http://user:pass@host:port.
| Variable | Default | Meaning |
|---|---|---|
SOLVEHUB_API_KEY | - | Console API key. Required for every tool except get_prices. |
SOLVEHUB_BASE_URL | https://api.tiam.codes | Override for a self-hosted or staging endpoint. |
SOLVEHUB_TIMEOUT_MS | 30000 | Per-request socket timeout. |
Idempotency-Key on every create, and retries 429/5xx and network errors with exponential backoff, so a retry cannot bill the same task twice.get_balance and get_prices are there so an agent can check the cost before spending anything.uncertain is not a failure: the challenge could not be judged yet and can still resolve, so wait_for_task keeps polling instead of giving up.| Type | Typical fields |
|---|---|
turnstile | websiteURL, websiteKey, optional action / cdata |
interstitial | websiteURL, proxy; returns the clearance cookie and user agent |
datadome | websiteURL, proxy, optional cookies |
akamai | websiteURL, proxy; returns _abck and bm_sz |
perimeterx | websiteURL, appId, proxy, cookies |
imperva | websiteURL, proxy, mode reese84 or UT.mvc |
aws_waf | challenge script URL, optional captcha script, proxy |
alibaba | scene id, identity prefix, verification payload |
funcaptcha | websiteURL, websiteKey, optional subdomain |
image | body with the base64 image; returns the recognised text |
Exact field names per type are in the API reference.
npm test
The tests run the JSON-RPC dispatcher, the tool schemas and the polling state machine against a mocked HTTP layer, so they need no API key and no network.
Use it on systems you own or have written permission to test. That is what the API is for, and the terms say the same thing.
MIT licensed.