# adityarya24/astro-skill [Health: Active]

**Category:** 🔮 Spirituality & Esoterica  
**Repository:** https://github.com/adityarya24/astro-skill  
**GitHub Stars:** 3  
**Views:** 5  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/adityarya24-astro-skill

## Description
Open-source Vedic astrology engine and stdio MCP server — deterministic kundali, Vimshottari dasha (maha/antar/pratyantar), panchang, gochar + Sade Sati, ashtakavarga, guna milan, divisional charts, and 12-page Hindi janma patrika PDFs. Bundled Swiss Ephemeris, SQLite client storage, no API key.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `uvx` (confidence: high):

```json
"mcpServers": {
  "astro-skill": {
    "command": "uvx",
    "args": ["astro-skill"]
  }
}
```

## Documentation

## What adityarya24/astro-skill MCP server does

The adityarya24/astro-skill MCP server provides an offline Vedic astrology engine and exposes it to MCP clients over stdio. It accepts birth details, dates, places, coordinates, and chart data, then returns structured calculations or generated reports. The engine uses bundled Swiss Ephemeris SWIEPH data through `pyswisseph` and Lahiri, also called Chitrapaksha, ayanamsa.

Its astrology output covers birth charts, Navamsa, planetary strengths, yogas, house analysis, Vimshottari dasha periods, daily panchang, planetary transits, Sade Sati, and Ashtakoot compatibility scoring. It also includes bilingual Hindi and English report generation through ReportLab. Calculations do not require an external API key, rate-limited service, or network connection.

## How it works

The MCP entry point runs as `astro-mcp`, or through `python -m services.astro_mcp`, and communicates with clients using stdio. Requests pass through the MCP layer to the shared astrology core, which returns structured JSON or creates a PDF report. The same core is also available through Python modules and command-line scripts, so an MCP workflow can share calculations with local automation.

The server includes SQLite-based client profile management. Agents can save birth parameters, search profiles by identifier or name substring, and retrieve reports associated with a client. This makes it suitable for repeated work with locally stored client records, while keeping the calculation and storage surfaces on the local machine.

## Setup and configuration

Install the package with Python 3.11 or newer:

```bash
pip install astro-skill
```

Start the stdio server with:

```bash
astro-mcp
```

A Python module invocation is also supported:

```bash
python -m services.astro_mcp
```

Claude Desktop can launch it with the `astro-mcp` command. Cursor can invoke `python -m services.astro_mcp` and set the repository path as its working directory. The provided setup does not require environment variables or API credentials.

## Tools and capabilities

The 11 MCP tools are:

- `parse_birth_details` converts conversational birth information into structured date, time, and coordinate fields.
- `save_client_profile`, `find_client_profile`, and `list_client_reports` manage local SQLite profiles and report records.
- `calculate_kundali` produces a Lahiri whole-sign chart with D1, D9, yogas, and bhava information.
- `calculate_dasha` calculates Vimshottari Mahadasha and Antardasha periods with birth-balance handling.
- `calculate_gochar` calculates transits and Sade Sati status for a selected date.
- `calculate_compatibility` returns a 36-point Ashtakoot Guna Milan result for two charts.
- `calculate_panchang` returns sunrise-based Tithi, Vara, Nakshatra, Yoga, and Karana data.
- `generate_report_json` combines astrology results into a structured draft.
- `generate_pdf_report` renders a bilingual printable report using in-process ReportLab.

## Limitations and notes

The adityarya24/astro-skill MCP server is described as a calculation and reporting engine, not as a general-purpose conversational astrology interpretation service. Its open-core repository provides mathematical calculations, JSON schemas, and factual PDFs; a separately described Pro Report Tier adds narrative synthesis, premium layouts, and prioritized remedies.

The server is specifically documented for stdio MCP use. The README names Claude Desktop, Cursor, and Codex among compatible MCP clients, but does not document an HTTP transport. Offline operation applies to the astrological calculations and does not imply that every surrounding client is offline.

_Full upstream README: https://allmcps.com/mcp/adityarya24-astro-skill/readme_