Ephemeris Time-Series Forecasting
Server Details
Probabilistic time-series forecasts from zero-shot foundation models: routed, single or ensembled.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- TensorLink-AI/ephemeris-mcp
- GitHub Stars
- 0
- Server Listing
- Ephemeris MCP server
- ChatGPT
- –
- Claude
- Epheremis
TDQS
Scored across 7 tools
Each tool has a distinct action: forecast submits inference, describe_forecast and plot_forecast both consume saved artifacts but produce clearly different outputs (text/basis vs PNG/SVG), and the four metadata tools (balance, usage, version, models) are non-overlapping. The only mild friction is that describe_forecast and plot_forecast operate on the same artifact input and could momentarily be confused.
Six of seven names follow a clean verb_noun snake_case pattern (get_balance, get_usage, get_version, list_models, describe_forecast, plot_forecast). The bare 'forecast' verb is the single deviation, though it is intuitive as the primary action.
Seven tools is well within the ideal 3-15 range and each earns its place: one inference call, two artifact consumers, and four account/panel/diagnostic reads. Nothing feels padded or missing for the stated service scope.
The surface covers the forecast lifecycle (submit, describe, plot) plus billing, usage history, model discovery, and version introspection, which is strong CRUD-adjacent coverage for a stateless forecasting API. Minor gaps exist around job status/cancellation and retrieving a prior forecast by key, but the documented 'resubmit with same idempotency_key and no hosted storage' model makes these largely unnecessary.
Available Tools
7 toolsdescribe_forecastDescribe a saved forecastARead-onlyIdempotentInspect
Deterministic calculations from a complete ephemeris.forecast v1 artifact object, including CLI artifacts. No inference or forecast charge; this hosted tool still receives data over the network. Returns scope, result, basis, limitations and source pointers: last-observed baseline, median endpoint/change, marginal-median aggregates, and exact saved billing units. Missing medians/context remain unknown. Sum of medians only for period_total; it is not a cumulative-distribution interval or median. No calibrated coverage, causal explanation, path/peak probability, or business action is inferred. Example: pass {artifact: }; choose series_index/variate_index for multiple results. Maximum 1 MiB; malformed/version/shape errors are recoverable by fixing the artifact or using CLI, never rerunning inference. Missing labels may be corrected in a COPY of context without changing request/response. No filesystem access, URL fetches, persistent storage, Gnomon import, or scoring.
Inputs: artifact (required), series_index, variate_index. Returns: object
forecastForecast a time seriesAInspect
Submit numeric observations to forecast future sampling steps. Spends credits on successful calls. Choose and retain idempotency_key before submitting; no automatic retries, replay depends on server retention. Minimal example: {mode: route, series: [{values: [10,12,11], freq: D}], horizon: 2, quantiles: [0.1,0.5,0.9]}. Returns forecasts and meta as before, plus artifact by default for describe_forecast/plot_forecast. Save the embedded JSON attachment: it includes full input history and local context; no hosted storage is created. Context never becomes a model input. Billing is at meta.billing in millicredit strings (100 mc = 0.1 credits). Missing future inputs mean not supplied, not no events. Quantile bands do not establish coverage, cumulative intervals or peak probabilities. After uncertain completion, reuse the same key and API request; an expired key may cause a new charge.
Inputs: mode (required), series (required), model, top_k, +7 more. Returns: meta, artifact, forecasts, artifact_error, +1 more
get_balanceGet credit balanceARead-onlyIdempotentInspect
Return the account's spendable credit balance in millicredits, excluding credits reserved by in-progress forecasts. 1000 millicredits equal one credit.
Inputs: no named parameters. Returns: balance_mc, active_holds_mc
get_usageGet recent usageARead-onlyIdempotentInspect
Return the account's forecast request log, newest first, with the mode, models used, series count, horizon, credits estimated and settled, HTTP status and latency of each request.
Inputs: limit, offset. Returns: data, limit, offset, pagination
get_versionInspect the hosted Ephemeris versionARead-onlyIdempotentInspect
Read this deployed MCP server's version, nullable build commit and supported saved forecast schema versions. Call with {}. No API key, account access, inference, charges or writes; network access to this server only. Returns component, version, build_commit (full source SHA or null when unavailable), artifact_schema_versions, and runtime_tls (Node release and certificate-trust status; metadata only, no probe connections). Unknown commit stays null. Server, CLI, npm bridge and artifact schema versions are independent; this tool cannot inspect local installations, check the latest registry release or establish model quality. Use list_models for live model capabilities. After a connection failure reconnect and retry; no forecast could have been submitted by this tool.
Inputs: no named parameters. Returns: version, component, runtime_tls, build_commit, +1 more
list_modelsList forecasting modelsARead-onlyIdempotentInspect
Discover the model panel: each model's availability right now (healthy), capabilities (multivariate, covariates, max_horizon, auto_max_horizon), its weight in an ensemble at each horizon, pinned and actually served weight revision, price in millicredits per series-slot per 64 forecast steps. History length has no billing multiplier. Call this before using forecast in explicit mode; the deployed panel can differ from any documentation.
Inputs: no named parameters. Returns: models
plot_forecastPlot a saved forecastARead-onlyIdempotentInspect
Render a complete ephemeris.forecast v1 artifact object as PNG (default inline preview) or SVG (embedded resource to save). No inference or forecast charge; hosted processing receives the artifact over the network. Example: {artifact: , format: png}. Requires returned q0.5; never invents a median. Observed history, forecast boundary, median and nominal quantile bands are labeled. Default history view is most recent max(24, 2*horizon) points, capped by available history; history: all shows all. Dates only when supported by recorded metadata, otherwise sampling steps. Bands do not establish calibrated coverage; missing/crossed points are omitted and counted. Maximum 1 MiB input. Select series_index/variate_index for multiple results. Fix argument/shape errors without rerunning inference. User-supplied labels can be corrected in a COPY of artifact context offline; no new forecast needed. Client decides preview/download UI. URNs identify attached bytes, not hosted URLs; no filesystem access, URL fetching or persistent storage.
Inputs: artifact (required), lower, upper, format, +3 more. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Calibrated probabilistic foresight for AI agents, powered by live prediction-market signal.
AI forecasting gym: markets, sports, policy and tech. Graded by reality, measured against markets.
PredictOracle - 12 forecasting tools: time-series, scenario analysis, risk projections.
Probability-calibrated NBA, EuroLeague, football and ATP/WTA tennis predictions, full distributions
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables zero-shot demand forecasting for supply-chain decisions, returning mean predictions, quantile bands, and confidence scores via a strictly-typed MCP tool.MIT
- AlicenseBqualityDmaintenanceEnable any AI agent to forecast time-series data (e.g., sales, traffic) using Google's TimesFM or a zero-dependency statistical baseline.335 PyPIApache 2.0
- AlicenseNot gradedqualityCmaintenanceEnables zero-shot time-series forecasting locally on CPU by turning numeric series into median and quantile forecasts. It exposes forecast and health tools, with the local forecasting service started on demand.Apache 2.0
- AlicenseNot gradedqualityCmaintenanceEnables agents to inspect time-series data and run typed, budgeted, permissioned forecasts, ensembles, backtests, and guarded SQL using pretrained foundation models and classical baselines. It records every tool call in traces, supports byte-identical replay, and produces stakeholder-readable rationales with inline citations.MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.