NEW: AI Agent Platform

More Than Chat. Build AI Astrology Agents

The only OpenAI-compatible agentic platform for astrology.

40 server-side tools, BYOK model choice, SSE streaming with automatic retry, multi-subject sessions. One API call, full agent loop.

View API Docs

LLM Context Presets

Prompt Languages

60%

Token Reduction

Agent Tools

OpenAI SDK Drop-In

Bring Your Own Key

Server-Side Agent Loop

See the Agent in Action

Real conversations powered by server-side tool orchestration

Astrology Chat APILive Demo

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.astrology-api.io/api/v3/chat",
  apiKey: "your-astrology-api-key",
});

const response = await client.chat.completions.create({
  model: "astro-default",
  stream: true,
  messages: [
    { role: "system", content: "Warm, insightful astrologer." },
    { role: "user", content: "Analyze my natal chart" }
  ],
  astrology: {
    subjects: [{ id: "me", birth_data: {
        year: 1990, month: 3, day: 15,
        hour: 14, minute: 30,
        city: "New York", country_code: "US"
      } }],
    enabled_tools: ["charts_natal"],
    defaults: { language: "en", tradition: "psychological" }
  }
} as any);

for await (const chunk of response) {
  process.stdout.write(chunk.choices[0]?.delta?.content || "");
}

Works with any of our 40 tools or combinations

Vedic Jyotish

Horary Questions

Human Design

Astrocartography

Numerology

Tarot Reading

Relationship Synastry

Transit Forecast

Any endpoint combination in one session — natal, Vedic, horary, BaZi, astrocartography, and more.Hosted: 25 + endpoint credits/turn · BYOK: 2 + endpoint credits + LLM costs

A full natal JSON dump eats 1,500-3,000 tokens per chart — context windows fill up fast
Cryptic field names like 'Mars 0d 90.0 Saturn' confuse the model and produce wrong readings
Transit and natal data mixed without prefixes — the LLM invents which planet is which
Locked into proprietary chat models with no BYOK choice or cost control
English-only prompt context blocks multilingual chatbot rollouts
Build your own tool loop, prompt formatter, and session manager from scratch

Transform with

to_prompt_text() formatter: 3 presets (minimal 150-250, full 500-800, traditional 600-900 tokens) — ~60% reduction vs raw JSON
Natural-language aspects: 'Mars square Saturn within 2°' instead of cryptic field codes
t. / n. prefixes disambiguate transit vs natal (e.g. 't.Jupiter trine n.Venus') — no more hallucinated planets
7-language i18n (~40 strings) covers EN, DE, ES, FR, IT, PT, RU for global chatbot deployment
BYOK: GPT-5, Claude 4, Gemini 3, Grok, or any of 50+ OpenRouter models
Server-side agent loop with 40 tools, automatic retry on upstream failures, multi-subject sessions, and binary artifacts

Platform Performance

Token-optimized context, agent-ready tooling, fast calculations

LLM Context Presets

Prompt Languages

~60%

Token Reduction vs JSON

~40

i18n Strings in Formatter

Agent Tools

<300ms

Calculation Speed

Interpretation Languages

50+

LLM Models

Complete Agent Platform Features

Everything you need to build AI astrology experiences

to_prompt_text() LLM Context Formatter

Bring Your Own Key (BYOK)

SSE Streaming & Events

40 Built-In Astrology Tools

Automatic Retry & Reliability

Multi-Subject & Sessions

Widget-Ready for AI Coding Tools

Binary Artifacts & Output

What Is Astrology Chat API? Complete Developer Guide

Everything you need to build AI-powered astrology experiences

What if you could add a fully-functional astrology AI agent to your app with a single API call? That's exactly what the Astrology Chat API does. It's not just a chatbot endpoint — it's a complete agentic platform that runs the entire tool-calling loop server-side. You send a user message, opt in the tools you want via astrology.enabled_tools, and the server calls into up to 40 astrology tools — natal charts, transits, synastry, full analysis reports, even PDF report generation — resolving birth chart data from subject IDs (no hallucinated dates) and returning a coherent response, all through an OpenAI-compatible /chat/completions endpoint. That means you can use the standard OpenAI SDK in TypeScript, Python, Go, or any language. No custom integration code. No building your own agent framework. No managing tool schemas. The server also classifies and retries transient upstream model failures automatically, so a flaky provider surfaces a clean error instead of a hung or garbled response. Existing astrology chat solutions charge $0.25-$0.87 per query, take 8-24 seconds to respond, lock you into proprietary models, and use custom JSON formats that require bespoke integration code. We're the only platform that ships a ready-made agent with BYOK support — choose your own LLM from GPT-5, Claude, Gemini, Grok, or any of 50+ models on OpenRouter.

How Does the Agentic Chat API Work?

The Astrology Chat API works like a specialized AI agent that happens to know astrology deeply.

Here's the flow: 1) You send a message via the standard OpenAI /chat/completions endpoint.

2) The server injects astrological context — your subjects' birth data, active tools, language preferences.

3) The LLM decides which tools to call (natal chart calculation, transit analysis, synastry report, etc.).

4) The server executes those tools internally against our 82 astrology endpoints — no extra API calls from your side.

5) Tool results are fed back to the LLM for synthesis.

6) You receive a coherent, grounded interpretation with real planetary data.

The entire tool loop runs server-side in a single request.

You never see tool-call messages — just the final answer.

With streaming enabled (SSE), you also get real-time extension events like tool_start and tool_end, so your UI can show 'Calculating transit chart...' while the agent works.

Entire agent loop runs server-side in a single API call

1 min read

What Is BYOK and Why Does It Matter?

BYOK — Bring Your Own Key — means you choose which LLM powers your astrology agent.

Every other astrology chat solution on the market locks you into a single proprietary model with no alternatives — some don't even disclose which model they use.

With our BYOK endpoint (/chat/completions/byok), you pass your own API key and model ID.

We support three providers: OpenAI (GPT-5, GPT-5-mini, GPT-5-nano, o3, o4), Anthropic (Claude Opus 4, Claude Sonnet 4, Claude Haiku 4), and OpenRouter (50+ models including Gemini 3, Grok 4, Llama, Mistral, and more).

Why does this matter? Cost control — GPT-5-nano is 10x cheaper than GPT-5-pro.

Quality tuning — Claude excels at nuanced interpretation, Gemini is faster for simple queries.

Compliance — some enterprises require specific model providers.

BYOK costs just 2 credits/turn (our platform fee) plus whatever your LLM provider charges.

50+ LLM models available — you choose cost, quality, and provider

1 min read

Why Is This the Only OpenAI-Compatible Option?

OpenAI compatibility means you use the standard OpenAI SDK — the same client library millions of developers already know.

No new SDK to learn, no custom HTTP client to build.

In TypeScript: new OpenAI({ baseURL: 'https://api.astrology-api.io/api/v3/chat' }).

In Python: OpenAI(base_url='...').

That's it.

Your existing code for ChatGPT works with minimal changes.

This also means instant compatibility with every framework built on OpenAI's protocol: LangChain, LlamaIndex, Vercel AI SDK, Semantic Kernel, AutoGen, CrewAI.

If your framework speaks OpenAI, it speaks Astrology Chat API.

No other astrology provider offers this — existing solutions use custom endpoints with proprietary request/response shapes, requiring bespoke integration code for every platform you support.

Works with any OpenAI SDK — TypeScript, Python, Go, Rust, and every AI framework

1 min read

Perfect for Vibe Coding — Build Astrology Widgets in Minutes

Because our API uses the standard OpenAI protocol, every AI coding tool already knows how to work with it.

Tell Lovable 'Build me an astrology chat widget using this API' and it generates a working React component — streaming UI, birth data form, chat bubbles, the works.

Claude Code and Cursor can scaffold a full integration in a single prompt because they understand the OpenAI SDK natively.

No custom adapters, no obscure documentation, no debugging proprietary JSON formats.

The same applies to Windsurf, Bolt, Replit Agent, and any AI tool that can write code against OpenAI's API.

This is what makes our platform uniquely suited for vibe coding: you describe the widget you want, the AI builds it, and it works on the first try — because the underlying API is the same protocol these tools were trained on.

Vibe coders are shipping astrology apps in hours, not weeks.

Chat widgets, compatibility checkers, daily horoscope generators, natal chart interpreters — all built by describing what you want to an AI coding assistant.

Copy this prompt into any AI coding tool to get started instantly:

Lovable, Claude Code, Cursor, Windsurf — any AI coding tool can build working integrations instantly

1 min read

What Can the 40 Agent Tools Do?

The platform exposes 40 tools the LLM can call mid-conversation, opt-in per request via astrology.enabled_tools — nothing runs unless you list it.

Chart Calculations (10 tools) — natal, transit, synastry, composite, davison, draconic, harmonic, heliocentric, local-space, and antiscia charts with full planetary positions and aspects.

Chart Rendering (11 tools) — the same chart set as SVG, PNG, JPG, WebP, or PDF images, stored as binary artifacts with shareable URLs.

Analysis Reports (4 tools) — full natal reports, lunar analysis, synastry reports, and relocation reports with multi-section interpretations.

PDF Report Generation (10 tools) — the agent can trigger a premium PDF mid-chat: archetypal map, composite, lunar return, multi-year roadmap, natal, relocation, saturn return, solar return, synastry, or Vedic Kundli report.

Data Lookups (3 tools) — current planetary positions, an aspect matrix, and the current astrological time.

Horoscope (1 tool) — personalized daily horoscope.

Rectification (1 tool) — birth time search.

Because tool names map directly to endpoint paths (/charts/natal becomes charts_natal, not natal_chart), a mistyped name is simply dropped — pass exact names and always attach subjects with birth_data, since tools resolve subjects by ID rather than reading birth data from the tool call itself.

40 tools, opt-in per request — the model can pull a real chart or generate a PDF report mid-conversation

1 min read

How Reliable Is the Chat API? Automatic Retry and Prompt Caching

A shipped reliability pass now classifies transient upstream model failures and retries them automatically, before they ever reach your app.

Previously a flaky upstream response could hang the connection or return a garbled partial reply; now a failure that can't be recovered surfaces as a clean 502 (non-streaming) or a terminal stream chunk (SSE) — your error handling gets a predictable signal instead of a silent stall.

That means meaningfully fewer broken chat sessions for anyone building on top of this endpoint.

A companion fix changed how the system prompt is assembled: it's now split into an always-on preamble plus a block of tool-usage rules that only loads when tools are actually resolved for that request.

This fixed a class of empty-completion failures that showed up disproportionately on lighter, faster model tiers, and it means prompt-cache behavior now tracks the tool set you actually enabled rather than a fixed static prompt — cheaper and more predictable costs for high-volume callers who vary enabled_tools across requests.

Automatic retry on transient upstream failures, plus caching that tracks the tools you actually enable

1 min read

How Do Multi-Subject Sessions and Artifacts Work?

Two features no competitor offers: multi-subject sessions and binary artifacts.

Multi-subject means you can pass multiple birth charts in a single conversation — for example, 'me' and 'partner'.

The LLM references subjects by their semantic ID when calling tools, and the server resolves them to full birth data.

This enables synastry analysis, composite chart comparisons, and family dynamics — all in one chat session, without re-sending birth data.

No other chat API supports more than one subject per conversation.

Binary artifacts let the chat generate downloadable files.

Ask 'Generate my natal chart' and receive an SVG chart image.

Ask 'Create a PDF report' and get a professional multi-page document.

Artifacts are stored in Supabase with shareable URLs and 24-hour retention.

Sessions persist via Redis with a 24-hour sliding TTL.

Pass the same session_id across turns and the server restores conversation history, subjects, and context — true multi-turn memory without client-side state management.

Multi-subject sessions and binary artifacts — both unique to our platform

1 min read

How to_prompt_text() Cuts LLM Token Costs by ~60%

Most developers ground their LLM chatbots by dumping the full chart JSON into the prompt.

A natal chart in JSON runs 1,500-3,000 tokens — that's $0.015-$0.030 per request on GPT-4-class models, plus the context window fills up after a few turns.

Worse, the model often misreads cryptic field codes like 'Mars 0d 90.0 Saturn' and produces wrong interpretations.

Our to_prompt_text() formatter solves both problems.

It converts chart data into compact, human-readable markdown with natural-language aspects.

Three presets let you pick the right token budget: 'minimal' (~150-250 tokens) covers sun/moon/asc, key aspects, current transits — great for chat replies.

'full' (~500-800 tokens) adds dignities, houses, modalities, and sect for richer interpretations.

'traditional' (~600-900 tokens) includes Hellenistic and medieval depth — lots, almuten, hayz, antiscia — for classical-style readings.

Drop the formatted markdown into any LLM prompt (Claude, GPT, Gemini, Grok) to ground the model in real chart data without dumping JSON.

Average token savings: ~60% versus a raw JSON dump, with measurably fewer hallucinations.

3 presets, 150-900 tokens, ~60% cheaper than JSON dumps

1 min read

Natural-Language Aspects and t./n. Prefixes

LLMs read natural language better than field codes — so to_prompt_text() outputs aspects the way an astrologer would write them.

Instead of 'Mars 0d 90.0 Saturn' you get 'Mars square Saturn within 2°'.

The orb stays precise, but the model parses it on the first pass.

When the formatter includes transit data alongside the natal chart, every planet is tagged with a prefix: 't.' for transit, 'n.' for natal.

So a transit reading shows 't.Jupiter trine n.Venus within 1°' — the model never has to guess which planet belongs to which chart.

This single change kills the most common source of hallucinated readings in astrology chatbots: planet identity confusion between charts.

Combined with the 7-language i18n (EN, DE, ES, FR, IT, PT, RU — ~40 strings covering aspects, sect, dignities, houses, modalities), you can ship a multilingual chatbot that grounds Claude or GPT in localized chart context without translating anything client-side.

Natural-language aspects + transit/natal prefixes eliminate hallucinated readings

1 min read

Performance and Pricing vs Competitors

Let's talk numbers.

The fastest competing astrology chat solutions take 8-12 seconds per response.

Standard-tier plans on the market average 19-24 seconds.

Our calculation layer responds in under 300ms — total response time depends only on LLM speed, typically 2-5 seconds with streaming.

That's 4-10x faster.

On pricing, alternatives charge $0.25-$0.87 per AI query — on top of monthly subscription fees.

Our pricing has three components: (1) platform credits per turn — 25 hosted or 2 BYOK, (2) endpoint credits for each tool the agent calls (e.g.

natal chart = 1 credit, astrocartography = 5 credits, PDF report = 35 credits), and (3) your own LLM costs if using BYOK.

Example: a natal interpretation turn costs 25 + 1 = 26 credits hosted (~$0.02), or 2 + 1 = 3 credits BYOK (~$0.002) plus ~$0.003 LLM cost.

That's 10-100x cheaper per conversation turn than alternatives.

We include chat in our standard free tier: 50 requests per month, no credit card required.

4-10x faster and 10-100x cheaper than competitors

1 min read

Use Cases

Astrology Mobile Apps

Add a conversational AI astrologer to your iOS or Android app. Users ask questions in natural language, get personalized interpretations backed by real chart calculations. OpenAI SDK works natively with React Native and Flutter.

Dating & Compatibility Platforms

Multi-subject sessions enable real-time compatibility analysis between two users. The agent calculates synastry, composite charts, and relationship dynamics — all in a single conversation flow.

Content Generation Pipelines

Generate personalized horoscopes, weekly forecasts, and natal reports at scale. BYOK with a cost-efficient model (GPT-5-nano) and batch process 2,800+ users for daily email or push notification content.

Wellness & Mental Health Platforms

Integrate astrological insights into mindfulness apps, therapy tools, and self-reflection platforms. The psychological tradition mode provides Jungian-style interpretations focused on growth and self-awareness.

SaaS Astrology Platforms

White-label the entire agent as your own product. Custom system prompts set the personality, BYOK controls costs, and session persistence enables multi-turn consultations for paying subscribers.

AI Agent Frameworks

Plug directly into LangChain, CrewAI, AutoGen, or any OpenAI-compatible agent framework. Our API becomes a specialized astrology tool in your multi-agent system — no custom adapters needed.

Vibe Coding with AI Tools

Tell Lovable, Claude Code, Cursor, or Windsurf to build you an astrology widget — and get a working app in minutes. Because our API follows the OpenAI protocol, AI coding tools generate correct integrations on the first try. Chat widgets, horoscope generators, compatibility checkers — describe it and ship it.

Technical Specifications

API Protocol	OpenAI Chat Completions (v1)
Endpoints	/chat/completions (hosted) + /chat/completions/byok
LLM Context Formatter	to_prompt_text() with 3 presets
Preset Values	minimal \| full \| traditional
Token Range per Preset	minimal 150-250 · full 500-800 · traditional 600-900
Aspect Format	'Mars square Saturn within 2°' (natural language)
Transit/Natal Prefixes	t.Jupiter trine n.Venus
Prompt Formatter Languages	7 (en, de, es, fr, it, pt, ru)
i18n Strings in Formatter	~40 across 7 languages
Agent Tools	40, opt-in per request via astrology.enabled_tools
Upstream Failure Handling	Automatic classification and retry, clean 502 or terminal stream chunk on failure
Calculation Speed	<300ms per tool call
Total Response Time	2-5s with streaming
Streaming	SSE with tool_start/tool_end extension events
BYOK Providers	OpenAI, Anthropic, OpenRouter (50+ models)
Hosted Model	astro-default (optimized for astrology)
Session Persistence	Redis-backed, 24h sliding TTL
Multi-Subject	Unlimited subjects per session
Binary Artifacts	SVG, PNG, JPG, WebP, PDF via chat
Interpretation Languages	9 (en, de, fr, es, it, pt, ru, zh, hi)
Traditions	Classical, Psychological, Event-Oriented, Universal
Authentication	Bearer token (API key)
Rate Limiting	Per API key, configurable

Astrology Chat API vs DIY JSON Dump Approach

Feature	Astrology API	Others
LLM Prompt Tokens (per chart)	150-900 (preset-dependent)	1,500-3,000 (raw JSON)
Token Cost per Prompt	~60% lower than JSON dump	Full token spend on every turn
Aspect Readability	'Mars square Saturn within 2°'	'Mars 0d 90.0 Saturn' (cryptic)
Transit/Natal Disambiguation	t. / n. prefixes built in	Model guesses (hallucinations)
Multilingual Prompt Context	7 languages (~40 i18n strings)	English-only or translate yourself
Hallucination Rate	Low — grounded natural language	High — cryptic codes confuse LLMs
OpenAI SDK Compatible	Yes (drop-in)	No (custom APIs)
BYOK Model Choice	50+ models (3 providers)	Locked proprietary model
Server-Side Agent Loop	40 tools, opt-in per request	Build it yourself
Upstream Failure Handling	Automatic retry + clean error signal	Silent hangs or garbled output
Multi-Subject Sessions	Unlimited subjects	Single birth chart
Binary Artifacts	PDF, SVG, PNG via chat	Not available
Response Time	2-5s (streaming)	8-24 seconds
Price Per Query	$0.001-$0.02	$0.25-$0.87
Free Tier	50 requests/month	None or paid sandbox
AI Coding Tools (Vibe Coding)	Works instantly (OpenAI protocol)	Custom formats, manual coding only

Related Astrology APIs

Natal Chart API

Calculate complete natal charts with planetary positions, houses, and aspects

Transit API

Current planetary transits — pairs with to_prompt_text() for chatbot context

Traditional Astrology API

Hellenistic and medieval techniques — matches the 'traditional' preset depth

Timing API

Electional and event timing — drop into chat prompts for forecast questions

Horary API

AI horary endpoint — pairs naturally with the chat agent for question-based readings

Synastry API

Relationship compatibility analysis between two charts

PDF Report API

Generate professional multi-page astrology PDF reports

API Documentation

Complete technical reference with interactive examples

References & Standards

Swiss Ephemeris (DE431-based) — astronomical calculations source
NASA JPL Planetary Ephemerides — underlying ephemeris dataset
International Astronomical Union (IAU) — body designation and nomenclature standards
IANA Time Zone Database — timezone & DST resolution

Copy this prompt into Lovable, Claude Code, Cursor, or any AI coding tool:

Build an astrology chat widget using the OpenAI SDK. Use base URL https://api.astrology-api.io/api/v3/chat and read the API docs at https://api.astrology-api.io/md/chat-completions for the full request schema. The widget should have: a birth data form (date, time, city), a streaming chat interface with message bubbles, and display tool call events (like 'Calculating natal chart...') as the agent works. Use the astrology namespace for subjects and enabled_tools as documented.

Simple, Transparent Pricing

Start free, scale as you grow. No hidden fees, no surprises.

Always Free

Start Free

Perfect for learning, testing, and small personal projects

50 API calls/month

Basic natal charts

5 house systems

Community support

Pro

$11/month

For small apps and personal projects

All basic endpoints
12 house systems
Email support

Best Value

Pro Plus

$21/month

For growing apps and medium traffic

All enhanced endpoints
18 house systems
AI interpretations
Priority email support

Ultra

$37/month

For professional astrology applications

Credit math ↓

All premium endpoints
23 house systems
Advanced AI interpretations
Priority support

High Volume Plans

For enterprise-scale applications

Business

$99/month

For high-volume applications

Credit math ↓

ALL endpoints
White-label ready
Custom AI fine-tuning
Phone support

UNLIMITED

Enterprise

$399+per month

Unlimited requests with dedicated infrastructure

Dedicated server infrastructure
Sub-300ms guaranteed
24/7 dedicated support
99.99% uptime SLA

All plans include comprehensive documentation and SDKs

Questions? Talk to our team

Frequently Asked Questions About Astrology Chat API

Real answers about the agentic astrology platform

General

What is the Astrology Chat API and how is it different from a regular chatbot?

Integration

Can I use the OpenAI Python or TypeScript SDK with this API?

Features

What LLM models are supported with BYOK?

Pricing

How much does each chat turn cost in credits?

Features

How does multi-subject chat work for synastry and compatibility?

Features

Can the chat generate PDF reports and SVG chart images?

Performance

How fast is the Astrology Chat API compared to alternatives?

Technical

What are the SSE streaming extension events?

Technical

How do sessions and conversation memory work?

Integration

What programming languages and frameworks are supported?

General

Is this just a GPT wrapper or does it use real astronomical data?

Features

How do I control the response style and length?

Features

What astrology traditions and house systems are supported?

Use Cases

Can I white-label this as my own product?

Technical

What happens if a tool call fails during the conversation?

Integration

Can I build an astrology widget with Lovable, Claude Code, or Cursor?

General

How does the Astrology Chat API compare to building my own with RoxyAPI or raw data endpoints?

Features

What is to_prompt_text() and why should chatbot developers care?

Features

Which preset should I use — minimal, full, or traditional?

Pricing

How much does the prompt formatter reduce LLM token costs?

Features

Which languages does to_prompt_text() support?

Technical

How are aspects formatted in natural language?

Technical

What do the t. and n. prefixes mean in the formatted output?

Features

Can the chat API pull my actual birth chart during a conversation?

Features

What tools can the AI call?

Technical

What happens if the AI model fails mid-response?

Pricing

Does tool use affect response caching or cost?

Build Your Complete Astrology Stack

The Chat API orchestrates these endpoints automatically, but you can also use them directly:

Ship Token-Optimized

OpenAI-compatible agent, 40 tools, BYOK, automatic retry on upstream failures, and to_prompt_text() context in 7 languages. Cut LLM token costs by ~60% versus dumping raw JSON.

Start Free View API Docs

50 free requests/month

OpenAI SDK compatible

No setup fees

More Than Chat. Build AI Astrology Agents

See the Agent in Action

Dumping Raw JSON Into LLM Prompts Wastes Tokens and Triggers Hallucinations

Token-Optimized LLM Context + Full Agent Platform in One API

Platform Performance

Complete Agent Platform Features

to_prompt_text() LLM Context Formatter

Bring Your Own Key (BYOK)

SSE Streaming & Events

40 Built-In Astrology Tools

Automatic Retry & Reliability

Multi-Subject & Sessions

Widget-Ready for AI Coding Tools

Binary Artifacts & Output

What Is Astrology Chat API? Complete Developer Guide

How Does the Agentic Chat API Work?

What Is BYOK and Why Does It Matter?

Why Is This the Only OpenAI-Compatible Option?

Perfect for Vibe Coding — Build Astrology Widgets in Minutes

What Can the 40 Agent Tools Do?

How Reliable Is the Chat API? Automatic Retry and Prompt Caching

How Do Multi-Subject Sessions and Artifacts Work?

How to_prompt_text() Cuts LLM Token Costs by ~60%

Natural-Language Aspects and t./n. Prefixes

Performance and Pricing vs Competitors

Use Cases

Astrology Mobile Apps

Dating & Compatibility Platforms

Content Generation Pipelines

Wellness & Mental Health Platforms

SaaS Astrology Platforms

AI Agent Frameworks

Vibe Coding with AI Tools

Technical Specifications

Astrology Chat API vs DIY JSON Dump Approach

Related Astrology APIs

Natal Chart API

Transit API

Traditional Astrology API

Timing API

Horary API

Synastry API

PDF Report API

API Documentation

References & Standards

Simple, Transparent Pricing

Always Free

Pro

Pro Plus

Ultra

Business

Enterprise

Frequently Asked Questions About Astrology Chat API

What is the Astrology Chat API and how is it different from a regular chatbot?

Can I use the OpenAI Python or TypeScript SDK with this API?

What LLM models are supported with BYOK?

How much does each chat turn cost in credits?

How does multi-subject chat work for synastry and compatibility?

Can the chat generate PDF reports and SVG chart images?

How fast is the Astrology Chat API compared to alternatives?

What are the SSE streaming extension events?

How do sessions and conversation memory work?

What programming languages and frameworks are supported?

Is this just a GPT wrapper or does it use real astronomical data?

How do I control the response style and length?

What astrology traditions and house systems are supported?

Can I white-label this as my own product?

What happens if a tool call fails during the conversation?

Can I build an astrology widget with Lovable, Claude Code, or Cursor?

How does the Astrology Chat API compare to building my own with RoxyAPI or raw data endpoints?

What is to_prompt_text() and why should chatbot developers care?

Which preset should I use — minimal, full, or traditional?

How much does the prompt formatter reduce LLM token costs?

Which languages does to_prompt_text() support?

How are aspects formatted in natural language?

What do the t. and n. prefixes mean in the formatted output?

Can the chat API pull my actual birth chart during a conversation?

What tools can the AI call?

What happens if the AI model fails mid-response?

Does tool use affect response caching or cost?