# Product capabilities

10 mechanisms, each described by observable behaviour, measured figures and its place in the product.

## Key facts

Stated as of 2026-10-18. Each line links the page that carries the fact in full.

- editions: Starter, Atelier and Portfolio, planned; none is on sale, and the site sells one thing, an optional refill of assistant credits ([join the early access](https://stochastly.com/join))
- AI add-on: Assistant + Agents, an optional subscription on any plan, 500 AI credits included each month
- access: no free plan; early access: free for life on one computer, in exchange for written feedback; an account and five short questions before the download
- platform: desktop application for Windows 10 or 11, x64, per-user installer downloaded from your account ([install guide](https://stochastly.com/docs/install))
- macOS: macOS 13.1 or later on Apple silicon; a .dmg from your account, signed with a Developer ID and notarised by Apple on 2026-10-16 ([install guide](https://stochastly.com/docs/install))
- Linux: no build, and no Linux direction on the roadmap
- version: 0.2.4, alpha; 25 dated entries from 2026-08-12, the first build, to 2026-10-18 ([version history](https://stochastly.com/version))
- live trading: through an exported file run where you trade (Pine Script on TradingView, MQL5 on MetaTrader, Python in your own setup); direct broker execution from the app is planned once the applicable regulation is in place ([roadmap](https://stochastly.com/roadmap))
- where it runs: on your computer: price files, strategies and results stay in the local Vault; the licence check sends the licence token, the device secret and a hashed machine fingerprint, and the optional hosted assistant sends its conversations to the Stochastly relay and on to Nebius B.V. in the Netherlands ([network traffic](https://stochastly.com/product/local) · [privacy](https://stochastly.com/privacy))
- publisher: Mahé Chadeffaud, sole proprietor trading as Stochastly, SIREN 109 270 363, Annecy, France ([legal notice](https://stochastly.com/legal)); founder and co-founder on the [credits](https://stochastly.com/credits) page

## What runs, and what is turned away

[Backtesting](https://stochastly.com/product/backtesting): the causality gate: bars truncated, graph replayed, a backend whose past moved is refused by name

[Runs locally](https://stochastly.com/product/local): a binary on your disk whose run output stays there, and the outbound hosts read out of the package by a static pass, one line per destination

[How it checks itself](https://stochastly.com/product/guards): validations that report missing prerequisites and replay their target failures

[Charts](https://stochastly.com/product/charts): every display of the Analyze workspace draws its own figure from the run on screen, the numeric readings folded under it

[Screenshot: Search landscape (contour) in Stochastly, Parameter search · 24 configs, light theme.](https://stochastly.com/screenshots-current/product--search-landscape-contour--search.png)

A contour view of the parameter-search results. Generated demonstration data, shown with the native chart controls.

## What you can take with you

[Exports](https://stochastly.com/product/exports): 6 targets compiled from one tree, each file stating what it leaves behind and its coverage measured on one synthetic price path

[Research papers](https://stochastly.com/product/papers): a typeset paper exported on request, whatever the verdict, with its disclosure box carrying the CFTC Rule 4.41 text verbatim

[Screenshot: Verdict at a glance in Stochastly, Full desk · 155 nodes, light theme.](https://stochastly.com/screenshots-current/product--verdict-at-a-glance--desk.png)

A graphical summary of the checks measured by the run; the recorded gate verdict remains NO-GO.

## The same pipeline, driven by an agent

[Assistant](https://stochastly.com/product/assistant): reads the run, opens the figure it cites, builds and launches a frozen /verdict experiment; each action is echoed on the desk

[Code editor](https://stochastly.com/product/editor): a Python file open on the desk, an edit proposed as a red and green diff, a custom node written and wired

[Research sessions](https://stochastly.com/product/research): several agents work one topic in turns; an idea is marked supported when a run or a note backs it, and the engine keeps the verdict

[MCP server](https://stochastly.com/product/mcp): an agent composes a graph and reads the verdict; a tool that changes something states what it changes [MCP launcher source on GitHubstochastly-mcp · MIT](https://github.com/Stochastly/stochastly-mcp)

## The catalogue, the agent surface, and what leaves the machine

- catalogue: 667 nodes · 28 families
- validated by executable checks: 492
- badged roadmap: 175
- neither validated nor marked for later work: 0
- agent surface: 35 MCP tools, each classed read or write · 6 resources
- strategy compiles to: cpp · mql5 · pine · python · rust, from one representation
- alpha expression exports to: FASTEXPR (WorldQuant BRAIN), the signal alone
- run output: stays on your disk unless you opt in to data sharing
- outbound calls: licence heartbeat (hardware id) · the assistant, when enabled · verdicts and run descriptions, if you ticked the two sharing boxes

measured 2026-09-16 — each count above is regenerated from the shipped catalogue

## Where the build is going, and what it carries today

[Version and roadmap](https://stochastly.com/roadmap): what the current milestone contains, feature by feature; then the directions under study, with what exists for each and no date attached

## A claim carries the file that settles it

Each mechanism above is described with its observable behaviour and limit. Figures are read from the shipped catalogue and carry their measurement date.

What a mechanism does not establish is written next to what it does. The validity domain of a measured result, with its perimeter and the control available in the build, is registered under Research.

## Product facts

Check plan rights, data responsibilities and execution limits.

---
Source: https://stochastly.com/product
