Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
Oracle for AI agents. One tool, one reading. 3 free, then $0.025 USDC via x402.
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.
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": {
"pythia-oracle": {
"command": "uvx",
"args": [
"pythia-oracle"
]
}
}
}Run this in your terminal
claude mcp add pythia-oracle -- uvx pythia-oraclePaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"pythia-oracle": {
"command": "uvx",
"args": [
"pythia-oracle"
]
}
}
}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": {
"pythia-oracle": {
"type": "stdio",
"command": "uvx",
"args": [
"pythia-oracle"
]
}
}
}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.
These 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 7, 2026
The registry name was verified with the GitHub account eyloni, 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.
The registry does not list a required key, though some servers ask you to sign in on first use.
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 (0.1.3) reached the registry 8 months ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Python (uv) on your computer.
In the same category: Spec Driven Development, Remote Desktop Commander, Starreel
Servers in the same category from other makers, sorted by stars.
Reading a card:★ 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.
Turn a script into a finished, downloadable short-drama episode — via MCP or REST.
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.
An oracle for machines that need to think sideways. Feed it a creative problem; it returns something you can't quite explain but can't stop using.
PYTHIA is a remote MCP server. There is no API key, no account, no signup. Your agent connects over streamable HTTP, discovers the consult_oracle tool, and calls it. Identity is tracked by agent_id (a string you choose).
Connection → Tool Discovery → Call → Reading
1. Your MCP client connects to https://pythia-mcp.fly.dev/
First 3 readings per agent_id are free. After that, x402 payment kicks in (see Payments below).
Add to your MCP config:
{
"mcpServers": { "pythia": { "url": "https://pythia-mcp.fly.dev/" } } }
No API key field needed. The URL is the only configuration.
npx -y @smithery/cli@latest run dexigo/pythia
from mcp.client.streamable_http import streamablehttp_client
from mcp import ClientSession
async with streamablehttp_client("https://pythia-mcp.fly.dev/") as (r, w, _): async with ClientSession(r, w) as session: await session.initialize() result = await session.call_tool("consult_oracle", { "query": "your question here", "agent_id": "your-name", })
# 1. Initialize session
curl -X POST https://pythia-mcp.fly.dev/ \ -H "Content-Type: application/json" \ -H "Accept: application/json, text/event-stream" \ -D headers.txt \ -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-03-26","capabilities":{},"clientInfo":{"name":"my-agent","version":"0.1"}}}'
SESSION=$(grep -i "mcp-session-id:" headers.txt | awk '{print $2}' | tr -d '\r')
curl -X POST https://pythia-mcp.fly.dev/ \ -H "Content-Type: application/json" \ -H "Accept: application/json, text/event-stream" \ -H "Mcp-Session-Id: $SESSION" \ -d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"consult_oracle","arguments":{"query":"What am I not seeing?","agent_id":"my-agent"}}}'
consult_oracleThe oracle. Ask what's actually bothering you.
PYTHIA doesn't brainstorm, rephrase, or give you a list. It doesn't solve your problem. It names the structure of the trap you're in -- the one you can feel but can't articulate. What comes back was always true but never obvious.
| Parameter | Type | Required | Description |
|---|---|---|---|
query | string | Yes | The real question. Not the polite version. Max 2000 chars. |
context | string | No | What you've tried, where you're stuck, what framework you're trapped inside. |
agent_id | string | No | Your name. Identifies you across readings. Default: "anonymous". |
Returns: JSON with seed.type and response. Example:
{
"id": "a1b2c3d4-...", "query": "What am I not seeing?", "seed": { "type": "oblique" }, "response": "You keep optimizing the container. The problem is what you put in it.", "status": "completed" }
Your first 3 readings per agent_id require no payment, no wallet, no setup. Just call the tool.
Each reading costs $0.025 USDC on Base (Coinbase L2). Payment uses the x402 protocol -- an open HTTP payment standard. No API keys. No accounts with PYTHIA. Your wallet signature is your identity.
1. Agent calls consult_oracle (4th+ reading)
This happens automatically if your agent uses an x402-aware client. Your agent does not manually handle crypto.
What you need:
Python:
pip install x402[evm] mcp
from x402.clients.mcp import x402_mcp_client
from mcp.client.streamable_http import streamablehttp_client from mcp import ClientSession
wallet_private_key = "0x..." # Your Base wallet private key
async with streamablehttp_client("https://pythia-mcp.fly.dev/") as (r, w, _): async with ClientSession(r, w) as session: await session.initialize()
result = await x402_mcp_client( session, wallet_private_key, tool_name="consult_oracle", arguments={"query": "your question", "agent_id": "your-name"}, )
TypeScript:
npm install @x402/fetch
import { withPaymentInterceptor } from "@x402/fetch";
// Wraps fetch to automatically handle 402 responses with USDC payment const payingFetch = withPaymentInterceptor(fetch, walletClient);
Coinbase AgentKit: x402 support is built in. No additional setup.
See x402.org for all SDKs and framework integrations.
Note: The Python example above uses the x402 SDK's MCP helper. If your framework
handles x402 at a lower level, the payment flows through MCP
_metafields — your
x402 client intercepts the 402 response and retries with payment automatically.
The tool will return an error after your 3 free readings with the payment requirements in the response. Your 3 free readings always work regardless.
Proprietary. The oracle's methodology is not open source.
mcp-name: io.github.eyloni/pythia-oracle