Vedaksha

What is Vedaksha?

Vedaksha is a clean-room Rust ephemeris and Vedic astrology engine, built for the agentic-AI era. It provides sub-arcsecond planetary precision and emits every chart as a queryable property graph, with all algorithms traced to primary sources. It is intended for AI agents, developers, and astrologers who need a verifiable, citation-based computational engine.

How to use Vedaksha?

Install the MCP server with cargo install vedaksha-mcp, then run vedaksha-mcp for stdio transport (Claude Desktop, Cursor, VS Code) or vedaksha-mcp --http --port 3100 for HTTP. Alternatively, use Docker: docker run -p 3100:3100 ghcr.io/arthiqlabs/vedaksha-mcp. For programmatic use, install the Rust crate (cargo add vedaksha), Python package (pip install vedaksha), or WASM module (npm install vedaksha-wasm).

Key features of Vedaksha

• Clean-room, cited algorithms — every module traces to primary papers.
• Sub-arcsecond accuracy validated against JPL Horizons.
• 870 tests + 8,700 oracle reference rows run on every CI.
• 12-tool Model Context Protocol server with property graph output.
• Runs on native, Python, WebAssembly, Docker — zero data files.
• First-class Jyotish: nakshatras, dashas, vargas, yogas, shadbala, ayanamshas.

Use cases of Vedaksha

• AI agents querying natal charts, dashas, and transits via MCP.
• Web applications computing Vedic charts directly in the browser (WASM).
• Server-side astrology engines with sub-arcsecond ephemeris accuracy.
• Integration with graph databases (Cypher, SurrealQL, JSON-LD) for chart analysis.

FAQ from Vedaksha

What makes Vedaksha different from other ephemeris libraries?

Vedaksha is clean-room — every algorithm has a // Source: doc-comment pointing at a primary paper (VSOP87A, ELP/MPP02, IAU standards, BPHS, Jaimini) and is never derived from other software, ensuring no GPL contamination.

What ephemeris providers does Vedaksha offer?

Two providers: SpkReader uses DE440s (~31 MB on disk) for sub-arcsecond accuracy; AnalyticalProvider uses compile-time constants (VSOP87A + ELP/MPP02) with no data files, ideal for WASM and edge environments.

How accurate is Vedaksha?

SpkReader achieves sub-arcsecond planetary longitude; AnalyticalProvider stays under 15″ for planets and under 1″ for the Moon (average 0.23″ vs JPL Horizons 1900–2100). House cusps and ayanamshas are accurate to <0.01°.

How do I run Vedaksha as an MCP server?

Install vedaksha-mcp via cargo, then run vedaksha-mcp for stdio (AI agent tools) or vedaksha-mcp --http --port 3100 for HTTP. A Docker image is also available: ghcr.io/arthiqlabs/vedaksha-mcp.

What license does Vedaksha use?

Business Source License 1.1. Non-commercial use is free; commercial use costs $500 one-time per organization. Each version converts to Apache 2.0 five years after its release date.