Vedic and Western astrology for AI agents: charts, dasha, matchmaking, panchanga, numerology, tarot.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent β or use 1-click editor setup below.
We haven't yet run this listing's install command through our automated sandbox check. This isn't a red flag β we're steadily working through the catalog.
π‘ Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Astrology and divination calculations as MCP tools. 103 tools covering Vedic and Western astrology, numerology, tarot, crystals, dreams, natal charts, Dasha, matchmaking and Panchanga, with interpretations that follow classical Jyotish method. Every position is verified against an independent Swiss Ephemeris run (asterwise.com/proof) and checked against NASA JPL Horizons, median 0.046 arcseconds over 80 positions (asterwise.com/accuracy).
See it first: 46-second demo of Claude Desktop casting a chart through this server, and the independent Swiss Ephemeris cross-check.
Sign up free at asterwise.com/dashboard: 500 calls/month on the Sandbox tier. No credit card. No time limit.
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
Add to .cursor/mcp.json in your project:
Three methods supported:
Method 1 β API Key (quick start)
Pass your Asterwise API key (starts with aw_) either as Authorization: Bearer <api-key> or as an X-API-Key: <api-key> header. Both are equivalent; use whichever your MCP client can set.
Method 2 β OAuth 2.1 (production)
Exchange your API key for a short-lived token:
Returns: {"access_token": "...", "expires_in": 3600, ...}
Use the token: Authorization: Bearer <access_token>
Access tokens are stateless HS256 JWTs. The API key is carried inside the token encrypted with a key derived from JWT_SECRET; only a SHA-256 hash of the key appears in the sub claim. Keep JWT_SECRET private.
Method 3 β OAuth 2.1 authorization code for MCP clients (Claude, Cursor, VS Code, Smithery)
MCP clients log a user in through the standard authorization-code flow with PKCE (S256). Discovery is at /.well-known/oauth-authorization-server. Two ways for a client to identify itself are supported:
client_id. The URL must serve a JSON document whose client_id equals the URL, with client_name, redirect_uris, and token_endpoint_auth_method: "none". These clients are public: no secret, PKCE only, authorization_code and refresh_token grants, refresh tokens rotate on every use. Redirect URIs must be https, or http on localhost / 127.0.0.1 / [::1] (port ignored, per RFC 8252). A reference document you can copy: https://asterwise.com/public/oauth/example-client.json.POST /oauth/register, deprecated in the MCP spec but still supported): returns a client_id and client_secret; the secret is required at /oauth/token.The server fetches metadata documents with SSRF protections (public addresses only, no redirects, 64 KB, 10 s) and caches them for the document's Cache-Control: max-age (60 s to 24 h, default 1 h). The consent page shows the host that published the document.
For hosts that speak stdio instead of HTTP (Glama hosted builds, a local Claude Desktop entry, quick tests):
Over stdio the key comes from ASTERWISE_API_KEY; the server starts and lists its tools without one, and tool calls need it. Logs go to stderr.
Copy .env.example to .env and set at least:
| Variable | Required | Description |
|---|---|---|
ASTERWISE_API_BASE_URL | Yes | Asterwise API base URL (e.g. https://api.asterwise.com). |
JWT_SECRET | For /oauth/token | At least 32 characters; used to sign access tokens. |
MCP_SERVER_HOST / MCP_SERVER_PORT | No | Bind address and port for the MCP HTTP server. |
LOG_LEVEL | No | Default INFO. |
MCP_OAUTH_SECRET | For OAuth | Shared with asterwise-api; verifies access tokens issued by its /v1/oauth/token. |
ASTERWISE_API_KEY | stdio only | Your Asterwise API key for python stdio.py; HTTP deployments take the key per request instead. |
INTERNAL_API_TOKEN | For OAuth client registration | Shared with asterwise-api; used when forwarding dynamic client registration. |
FRONTEND_URL | For /authorize | Where the browser is sent for sign-in and consent (e.g. https://asterwise.com). |
OPENAI_APPS_CHALLENGE_TOKEN | No | Served at /.well-known/openai-apps-challenge for directory verification. |
The MCP server exposes 103 tools organized by Python module. The categorization reflects code organization; tools may serve multiple traditions (e.g. matchmaking includes both Sanskrit Dashakoot and Tamil Porutham methods).
For the full tool list see docs.asterwise.com or the MCP server's tool listing endpoint.
uvicorn server:app is the same entry point the production Dockerfile and railway.toml use, so the auth middleware, OAuth routes and /health are all present locally. Runtime dependencies are pinned in requirements.txt; requirements-dev.txt adds the test tooling.
Coverage is enforced at 78% for core modules (auth, client, errors, logging_config, models, runtime, server); tool modules are excluded from the gate (see .coveragerc).
MIT. See LICENSE. Security reports: see SECURITY.md.
No reviews yet β be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/asterwise-2)<a href="https://allmcps.com/mcp/asterwise-2"><img src="https://allmcps.com/api/badge/asterwise-2?style=directory" alt="Asterwise on AllMCPs" /></a>