Skip to content

quickstart.mdx

Latest commit

 

History

History
151 lines (102 loc) · 3.72 KB

quickstart.mdx

File metadata and controls

151 lines (102 loc) · 3.72 KB
title Quickstart
description Get started with Timepoint AI — render your first historical moment in minutes.

Quickstart

Hosted API — The Fastest Path

The hosted API at api.timepointai.com is the easiest way to use Timepoint. The Gateway handles authentication, credits, and proxies your requests to Flash and other services.

Authenticate

Sign up through the Timepoint app to get your credentials. The Gateway supports Apple, Google, and GitHub OAuth.

Render a Moment

curl -X POST https://api.timepointai.com/api/v1/timepoints/generate/sync \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_JWT_TOKEN" \
  -d '{"query": "AlphaGo plays Move 37, Seoul, March 10 2016", "generate_image": true}'

You'll get back a complete scene: characters with distinct voices, period-accurate dialog, relationship dynamics, source citations, and a photorealistic image.

Query the Clockchain

The Clockchain is a public API — no authentication needed for read access.

# Get graph statistics
curl https://clockchain.timepointai.com/api/v1/stats

# Search moments by keyword
curl "https://clockchain.timepointai.com/api/v1/moments?q=rome"

Connect via MCP

The Clockchain exposes a Model Context Protocol endpoint for LLM tool use:

Endpoint: https://clockchain.timepointai.com/mcp/
Transport: Streamable HTTP (v1.26.0)

Point any MCP-compatible client (Claude Desktop, Cursor, etc.) at the endpoint to give your LLM direct access to the temporal causal graph.

Local Development

The sections below cover running Timepoint services locally from source.

Flash — Render a Historical Moment

Flash is a pure generation engine. In production, the Gateway handles auth and proxies requests to Flash. Locally, you can run it standalone.

Prerequisites

Install and Run

git clone https://github.com/timepointai/timepoint-flash.git
cd timepoint-flash
pip install -e .

Set your API key:

export GOOGLE_API_KEY="your-key-here"

Start the server:

flash serve

Render a Moment

curl -X POST http://localhost:8000/api/v1/timepoints/generate/sync \
  -H "Content-Type: application/json" \
  -d '{"query": "AlphaGo plays Move 37, Seoul, March 10 2016", "generate_image": true}'

Pro — Run a Social Simulation

Pro is the SNAG engine — it simulates social dynamics with causal provenance.

Install and Run

git clone https://github.com/timepointai/timepoint-pro.git
cd timepoint-pro
pip install -e .

Set your LLM provider:

export OPENROUTER_API_KEY="your-key-here"

Run a Simulation

bash run.sh --template board-meeting --fidelity 3

Pro supports 21 verified templates at $0.15–$1.00 per run.

Clockchain — Query the Graph Locally

See the Hosted API section above for querying the public Clockchain. To run locally:

git clone https://github.com/timepointai/timepoint-clockchain.git
cd timepoint-clockchain

Refer to the repository README for local setup instructions.

Next Steps

Understand timepoints, SNAG, and temporal modes The 14-agent pipeline explained Full endpoint documentation How the temporal graph works