Ephemeris Time-Series Forecasting

industries.cascadev1.0.1Updated Oct 6, 2026

Probabilistic time-series forecasts from zero-shot foundation models: routed, single or ensembled.

VerifiedStreamable HTTPWeb executableAI & MLData & Analytics

Overview

AI-generated overview

Lets an assistant produce probabilistic time-series forecasts with prediction intervals from historical numeric data, using hosted zero-shot forecasting models.

What it does
Exposes a forecast tool that takes historical numeric series and returns quantile forecasts, supporting one to 64 series per call, horizons up to 512 steps, optional covariates, and a choice of routing, ensemble, or explicit model selection. Additional tools list the live model panel with health, capabilities, horizon limits, ensemble weights and prices, and report spendable credits and recent request costs. Forecasting is zero-shot: no training, feature engineering, or GPU is required.
When to use it
Useful when an assistant needs to project numeric series such as sales, demand, inventory, traffic, signups, revenue, energy load, prices, or sensor and infrastructure metrics, and the user wants prediction intervals rather than a single point estimate. It fits workflows where several series must be forecast in one call or where model choice should be routed or ensembled automatically.
Requirements
A remote Streamable HTTP endpoint or the local npm package ephemeris-mcp run via npx for stdio-only clients. An Ephemeris account with credits and an API key is required, supplied as the EPHEMERIS_API_KEY environment variable or as an Authorization bearer token header. Network access to the service is needed.
Before you install
The service is paid per forecast and requires an API key (EPHEMERIS_API_KEY or the Authorization header) that grants spending against account credits; the key should be kept out of shared configuration. Historical series sent for forecasting leave the user's machine and go to the provider. The tools are read-only with respect to user data, but usage and balance queries expose account spending information.

Installation

In SourceWeft

  1. Open Ephemeris Time-Series Forecasting in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "ephemeris": {
      "type": "http",
      "url": "https://ephemeris.cascade.industries/api/mcp"
    }
  }
}

README

Ephemeris MCP server: time-series forecasting for AI agents

[Add Ephemeris to Cursor]

Give Claude, Cursor, ChatGPT or any MCP client the ability to forecast numeric time series with prediction intervals: sales, demand, inventory, web traffic, signups, revenue, energy load, prices, sensor readings, infrastructure metrics.

Ephemeris runs a panel of open-weights, zero-shot forecasting foundation models behind one API key:

ModelPublisherUse by name
Chronos-2Amazonchronos2
TimesFM 2.5Google Researchtimesfm25
Toto 2Datadogtoto2-313m
TiRex-2NXAItirex2
PatchTST-FM r2IBM Granitepatchtst-fm-r2
FlowState r1IBM Graniteflowstate-r1

Send history, get quantile forecasts back. No training, no feature engineering, no GPU. Name a model, let Ephemeris route to the best fit for your data, or use the ensemble, an accuracy-weighted blend of the panel:

  • TIME: level with the top of the leaderboard (MASE 0.639 vs 0.638 for the leader), with the best average MASE rank of 31 models
  • GIFT-Eval: CRPS 0.4662 against seasonal naive, ahead of every open-licence model

Scored with each benchmark's own harness. Details: ephemeris.cascade.industries/benchmarks.

Tools

ToolWhat it does
forecastForecast 1 to 64 series in one call: route, ensemble or explicit mode, any quantiles, optional covariates, horizons up to 512 steps
list_modelsThe live panel: health, capabilities, horizon limits, ensemble weights, prices
get_balanceSpendable credits
get_usageRecent requests and what each cost

Get an API key

Sign up at ephemeris.cascade.industries, add credits, and create a key (pc_live_...) in the dashboard. Pay per forecast, no subscription: pricing.

Connect

Remote server (Streamable HTTP): https://ephemeris.cascade.industries/api/mcp, header Authorization: Bearer pc_live_...

Claude Code (plugin: MCP server plus a forecasting skill)

/plugin marketplace add TensorLink-AI/ephemeris-mcp/plugin install ephemeris@ephemeris

You are asked for your API key once; it is stored in your system's secure credential store.

Claude Code (server only)

claude mcp add --transport http ephemeris https://ephemeris.cascade.industries/api/mcp \  --header "Authorization: Bearer pc_live_your_key"

Cursor: one click with [Add to Cursor], then replace YOUR_EPHEMERIS_API_KEY with your key in Cursor's MCP settings. Or add it by hand:

Cursor (.cursor/mcp.json) and most clients

json
{  "mcpServers": {    "ephemeris": {      "url": "https://ephemeris.cascade.industries/api/mcp",      "headers": { "Authorization": "Bearer pc_live_your_key" }    }  }}

VS Code (.vscode/mcp.json)

json
{  "servers": {    "ephemeris": {      "type": "http",      "url": "https://ephemeris.cascade.industries/api/mcp",      "headers": { "Authorization": "Bearer pc_live_your_key" }    }  }}

Claude Desktop and other clients that only run local (stdio) servers

json
{  "mcpServers": {    "ephemeris": {      "command": "npx",      "args": ["-y", "ephemeris-mcp"],      "env": { "EPHEMERIS_API_KEY": "pc_live_your_key" }    }  }}

OpenAI Responses API, Anthropic Messages API, Codex, Gemini CLI: see the docs.

Try it

Once connected, ask:

  • "Here are my last 18 months of sales: … Forecast the next 6 months with an 80% interval."
  • "Forecast next week's hourly traffic from this CSV and tell me the likely peak."
  • "Use the ensemble to project daily signups for 90 days; plot the median and the 10th to 90th percentile band."

More in examples/prompts.md. Without MCP, the same forecast is one REST call: examples/rest_forecast.py.

Reference

The code in this repository (the plugin manifest, skill and stdio bridge) is MIT-licensed. The models keep their own licences, listed on each model page.

Source: README.md at commit 647576f

Tools

0
Tool metadata has not been indexed yet.

Version history

1
  1. v1.0.1LatestOct 6, 2026