[Skip to content](#main-content) [![Jx Suite](/favicon.svg) Suite](/)Search⌘KESC Jump to [Install Jx Studio](/docs/start/install)[Build your first project](/docs/start/first-project)[Document format](/docs/framework/concepts/documents)[Formula catalog](/docs/framework/reference/formulas) ↵ select↑↓ navigateesc close [Studio](/studio)[Framework](/framework)[Templates](/templates)[Compare](/compare)[Docs](/docs/start/first-project)[GitHub](https://github.com/jxsuite/jx)[Download](/download) ✕ [Studio](/studio)[Framework](/framework)[Templates](/templates)[Compare](/compare)[Docs](/docs/start/first-project)[GitHub](https://github.com/jxsuite/jx) Color scheme [Download](/download) ☰ Docs menu Start here[Overview](/docs/start/)Get started[Install Jx Studio](/docs/start/install/)[Your first project](/docs/start/first-project/)[A tour of Jx Studio](/docs/start/studio-tour/)Tutorials[Tutorial: interactive component](/docs/start/first-component/)[Tutorial: blog with collections](/docs/start/first-collection/)[Video tutorials](/docs/start/videos/)Coming from another tool[Coming from WordPress](/docs/start/coming-from-wordpress/)[Coming from Webflow or Framer](/docs/start/coming-from-webflow/)[Coming from Wix or Squarespace](/docs/start/coming-from-wix/)[Coming from Lovable or Builder.io](/docs/start/coming-from-ai-builders/)Studio[Overview](/docs/studio/)[The desktop app](/docs/studio/desktop/)The workspace[Overview](/docs/studio/interface/)[Welcome screen](/docs/studio/interface/welcome-screen/)[Editors and views](/docs/studio/interface/modes/)[The canvas](/docs/studio/interface/canvas/)[Documents and panes](/docs/studio/interface/tabs/)[Quick Access](/docs/studio/interface/quick-access/)[Keyboard shortcuts](/docs/studio/interface/shortcuts/)[Preferences](/docs/studio/interface/preferences/)[Languages](/docs/studio/interface/languages/)[Problems and progress](/docs/studio/interface/problems-and-progress/)[Commands](/docs/studio/interface/commands/)Projects[Overview](/docs/studio/projects/)[Create a project](/docs/studio/projects/create/)[Pages, layouts, components](/docs/studio/projects/pages-layouts-components/)[The Content workspace](/docs/studio/projects/browse/)[Starter templates](/docs/studio/projects/starters/)[Media](/docs/studio/projects/media/)[Content types](/docs/studio/projects/content-types/)[Project settings](/docs/studio/projects/settings/)[Dependencies and imports](/docs/studio/projects/dependencies/)Edit mode[Overview](/docs/studio/editing/)[Writing and formatting](/docs/studio/editing/writing/)[Slash commands](/docs/studio/editing/slash-commands/)[Frontmatter and metadata](/docs/studio/editing/frontmatter/)[Grid mode](/docs/studio/editing/grid/)Design mode[Overview](/docs/studio/design/)[Outline](/docs/studio/design/layers/)[Insert](/docs/studio/design/elements/)[Content tab](/docs/studio/design/properties/)[Style inspector](/docs/studio/design/style-inspector/)[Breakpoints](/docs/studio/design/breakpoints/)[States and selectors](/docs/studio/design/states-and-selectors/)[Design tokens](/docs/studio/design/tokens/)[Project Styles](/docs/studio/design/stylebook/)[Working with components](/docs/studio/design/components/)[Repeaters](/docs/studio/design/repeaters/)Logic[Overview](/docs/studio/logic/)[Formulas and expressions](/docs/studio/logic/formulas/)[Formula workspace](/docs/studio/logic/formula-workspace/)[Statements](/docs/studio/logic/statements/)[Logic tab](/docs/studio/logic/events/)[Data](/docs/studio/logic/data/)[Data sources](/docs/studio/logic/data-sources/)[Code editing](/docs/studio/logic/code/)Databases[Overview](/docs/studio/data/)[Connections](/docs/studio/data/connections/)[Data tables](/docs/studio/data/tables/)[Data grid](/docs/studio/data/grid/)[Auth and secrets](/docs/studio/data/auth-and-secrets/)AI assistant[Overview](/docs/studio/ai/)[The AI assistant](/docs/studio/ai/chat/)[Document assistant](/docs/studio/ai/document-assistant/)Publish[Overview](/docs/studio/publish/)[Source control](/docs/studio/publish/source-control/)[GitHub](/docs/studio/publish/github/)[Cloudflare Pages](/docs/studio/publish/cloudflare/)[Other hosts](/docs/studio/publish/other-hosts/)[Redirects](/docs/studio/publish/redirects/)[Real-time collaboration](/docs/studio/publish/collaboration/)Framework[Overview](/docs/framework/)Concepts[Documents](/docs/framework/concepts/documents/)[Elements](/docs/framework/concepts/elements/)[Components](/docs/framework/concepts/components/)[Props and scope](/docs/framework/concepts/props-and-scope/)[State](/docs/framework/concepts/state/)[Reactivity](/docs/framework/concepts/reactivity/)[References](/docs/framework/concepts/references/)[Expressions](/docs/framework/concepts/expressions/)[Statements](/docs/framework/concepts/statements/)[Functions and sidecars](/docs/framework/concepts/functions/)[Styling](/docs/framework/concepts/styling/)[Color schemes](/docs/framework/concepts/color-schemes/)[Popovers and overlays](/docs/framework/concepts/overlays/)[Accessibility](/docs/framework/concepts/accessibility/)[Lists and iteration](/docs/framework/concepts/lists/)[Data prototypes](/docs/framework/concepts/data-prototypes/)[Timing](/docs/framework/concepts/timing/)[Dynamic switching](/docs/framework/concepts/switching/)[Security](/docs/framework/concepts/security/)Site architecture[Overview](/docs/framework/site/)[project.json](/docs/framework/site/project-json/)[Routing](/docs/framework/site/routing/)[Layouts](/docs/framework/site/layouts/)[Content collections](/docs/framework/site/content-collections/)[Jx Markdown](/docs/framework/site/jx-markdown/)[Obsidian vault as content](/docs/framework/site/obsidian-vault/)[Relationships](/docs/framework/site/relationships/)[SEO and metadata](/docs/framework/site/seo/)[Locales and languages](/docs/framework/site/i18n/)[Images](/docs/framework/site/images/)[Redirects](/docs/framework/site/redirects/)[Feeds](/docs/framework/site/feeds/)[Site search](/docs/framework/site/search/)[Build output and adapters](/docs/framework/site/deployment/)How compilation works[Overview](/docs/framework/build/)[The dev server](/docs/framework/build/dev-server/)[CLI commands](/docs/framework/build/cli/)Working with agents[Overview](/docs/framework/agents/)[Authoring rules for agents](/docs/framework/agents/authoring-rules/)[Machine-readable docs](/docs/framework/agents/machine-readable/)Reference[Overview](/docs/framework/reference/)[Formula catalog](/docs/framework/reference/formulas/)[Operator reference](/docs/framework/reference/operators/)Extending[Overview](/docs/extending/)Extensions[Extension anatomy](/docs/extending/extensions/anatomy/)[Schema composition](/docs/extending/extensions/schema-composition/)[Custom classes](/docs/extending/extensions/classes/)[Custom formats](/docs/extending/extensions/formats/)[Capability methods](/docs/extending/extensions/capabilities/)[Project sections](/docs/extending/extensions/project-sections/)[Server mounts](/docs/extending/extensions/server/)[Connectors](/docs/extending/extensions/connectors/)[Search indexes](/docs/extending/extensions/search/)[Security and secrets](/docs/extending/extensions/security/)[Tutorial: a TOML format](/docs/extending/extensions/tutorial-toml-format/)[Tutorial: a guestbook](/docs/extending/extensions/tutorial-guestbook/)[First-party extensions](/docs/extending/extensions/first-party/)Embedding[Overview](/docs/extending/embedding/)[Hosting the Studio assets](/docs/extending/embedding/hosting/)[Writing a platform adapter](/docs/extending/embedding/platform-adapter/)[The backend protocol](/docs/extending/embedding/backend-protocol/)[Dev server internals](/docs/extending/embedding/dev-server/)[Embedding the runtime in your app](/docs/extending/embedding/runtime-host/)[Writing assistant tools](/docs/extending/embedding/assistant-harness/)UI kit[The Jx UI kit](/docs/extending/ui-kit/)Reference[Protocol routes](/docs/extending/reference/studio-routes/)[Implementation status](/docs/extending/reference/implementation-status/)[Spec changelog](/docs/extending/reference/spec-changelog/)[Standards alignment](/docs/extending/reference/standards/)Contributing[Working in the monorepo](/docs/extending/contributing/monorepo/)[Contributing to these docs](/docs/extending/contributing/docs/)[AI assistant evals](/docs/extending/contributing/ai-evals/) # AI assistant evals Studio's [AI assistant](/docs/studio/ai) is not one thing you can unit-test. It is a model you do not control wrapped in scaffolding you do: a system prompt, a set of tool schemas, the translations that turn a validation error back into an instruction the model can act on. Change any of that and the only honest question is whether the assistant got better, which a test suite cannot answer and a hunch should not. `packages/studio/evals/` is the harness that answers it. Reach for it whenever you touch `ai-system-prompt.ts`, `ai-tools.ts`, or the few-shot examples. ## The invariant **The model is held fixed; only the scaffolding is iterated.** That is the whole design. Because the model, the tasks, and the graders stay constant between runs, a difference in pass-rate is attributable to the diff you just made. Change two things at once (the prompt and the model, say) and the run tells you nothing. The second half of the design is that the harness drives the **real** production loop: the actual `runAgentLoop`, the actual `@jxsuite/ai` tool registry, the actual system prompt built by `buildSystemPrompt`. The only substitution is the LLM client, where the fake test client is swapped for a real OpenAI-compatible one. There is no parallel "eval mode" implementation to keep in sync, so a scaffolding change is exercised exactly as users will meet it. ## Golden tasks Each task in `evals/tasks/*.json` is one isolated, unambiguous specification: ```json shiki ``` `initialDoc` is the document the assistant starts from. Every trial gets a fresh tab, so nothing leaks between runs. `intent` records the human success criteria: it is what you read transcripts against today, and the hook for a future LLM-as-judge grader. Keep the suite stable. Tasks are the benchmark, and a benchmark that moves with the code cannot measure it. A task can also start with **no document open**, which is how the assistant works when a project is open and nothing is on the canvas. Such a task has no `initialDoc`; instead `files` is the project (each project-relative path mapped to its text, served from memory) and `output` names the file the trial is graded on. The model gets the same project-mode prompt and the same gated file tools (`list_files`, `read_file`, `write_file` and the rest) the assistant would offer, and a trial that never writes `output` fails. `about-page-no-document` and `footer-from-header-no-document` are the two in the suite. ## Running it ```bash shiki ``` `OPENAI_BASE_URL` and `OPENAI_MODEL` (default `gpt-4o`) are optional and mirror how the server proxy resolves its config. Without `OPENAI_API_KEY` the CLI exits `2` immediately, because the harness calls a real model by design. Any OpenAI-compatible endpoint works, Cloudflare Workers AI included: point `OPENAI_BASE_URL` at `https://api.cloudflare.com/client/v4/accounts//ai/v1`, pass an API token as the key, and name a model that calls tools, such as `@cf/meta/llama-4-scout-17b-16e-instruct`. The harness reaches the provider the way a server does, so an endpoint that answers no browser preflight is fine. If no trial gets a reply from the model at all (a bad key, an endpoint that is down or refuses the request, or a model that answers every trial empty), the CLI exits `2` and writes no run, because a run like that measures the endpoint rather than the assistant, and it would become the baseline the next run is compared against. Each run writes a timestamped directory under `evals/runs/`, which is gitignored: - `report.md` is the human summary: mean pass-rate, pass\@k and pass^k counts, the run's token totals with the share of input served from the provider's prompt cache, and a per-task table with `Δrate` against the previous run. - `results.json` holds the same numbers, machine-readable, minus the transcripts. - `transcripts/-.md` is one file per trial: the round and tool-call counts, both graders' verdicts, the final document, and the full message transcript. The token figures are the provider's own: each trial sums the `usage` frames its requests streamed. The client is created with a prompt-cache key of `eval--`, so a trial's rounds are routed to the same machine and can reuse what the earlier rounds cached. The key is a routing hint, not an isolation boundary: providers keep one prefix cache per machine, every trial sends the same reference material, and a different key can still land on a machine that has it. So the ratio can include hits from an earlier trial, and, because the key is the same in every run, from a run started within the provider's cache lifetime (minutes, up to about an hour). The cache ratio is still the number to watch after a system-prompt change: the prompt's reference material is most of every request, so a change that puts something dynamic ahead of it shows up as a lower ratio before it moves any pass rate. Compare runs that started from a cold cache, or both from a warm one, so cross-run warmth does not stand in for a layout change. A provider that reports no counts leaves the ratio at `n/a`. **Read the transcripts.** A grader you have not watched is a grader you cannot trust, and the failure that matters is usually visible in the third assistant turn rather than in the summary row. ## The two graders **Render critic: the primary signal.** It mounts the produced document with the real `@jxsuite/runtime` under happy-dom, the same render path the Studio canvas uses, and fails on anything the runtime surfaces: a thrown error during scope-building or node rendering, or a `console.error`/`console.warn`. In practice that means unresolved `$ref`s, missing `$prototype` classes, and broken bindings. Errors are phrased as actionable corrections ("→ Fix: a template binding references state that doesn't exist"), so the same output could later be fed back into the live loop. **Schema grader: the baseline.** The same `validateDoc()` the agent loop already self-corrects against. It is free and deterministic, so it is reported alongside for context. A trial passes on the **render critic**, not the schema grader. A document can be perfectly schema-valid and still render nothing useful, which is precisely the gap the critic exists to close. A trial whose loop failed before the model answered at all does not pass either way: its document is the task's untouched starting point, which renders, so grading it would score a failed request as a success. ## pass\@k and pass^k Models are non-deterministic, so a single trial per task measures luck. Each task runs `k` times (3 by default) and the scoreboard reports both: - **pass\@k** means at least one of the `k` trials passed. Capability: can the assistant do this at all? - **pass^k** means all `k` trials passed. Reliability: can a user count on it? `passRate` (passes ÷ k) sits between them and is what the mean and the `Δrate` column are computed from. ## The merge rule 1. Run `bun run eval` and note the baseline mean pass-rate; collect the failing transcripts. 2. Read the failures, then change **one** scaffolding file: the system prompt, the tool schemas plus `translateValidationError`, or the few-shot examples. One file per experiment keeps runs comparable. 3. Re-run. **Keep the change only if the mean pass-rate improves and `regressed` is empty.** Otherwise discard it. The scoreboard computes `regressed` by diffing against the most recent prior run: any task that passed pass\@k before and does not now. The CLI exits `1` when that list is non-empty, so the same command can gate CI. Warning Never tune the tasks to make a scaffolding change look good. If a task is genuinely ambiguous, fix the task in its own commit and re-baseline, but treat that as a change to the benchmark, not a result. ## The headless rubric harness `bun run eval:headless` runs a second harness at `packages/studio/tests/harness/`, which landed alongside the one above. It drives a catalog of prompts through the same production loop and scores them on rubric axes with evidence attached, running each test three times (`JX_AI_RUNS`) and reporting the **worst** run per axis so a borderline result cannot cherry-pick a lucky pass. It reads its key from `JX_AI_KEY`, falling back to `OPENAI_API_KEY`, and loads the repo-root `.env` regardless of the working directory. Use it for a qualitative read on assistant behavior; use `bun run eval` for the pass/fail number you gate on. ## Out of scope The harness deliberately does not do runtime UX sensors in the live assistant, LLM-as-judge grading, or autonomous self-editing. The render critic's error format is LLM-ready on purpose, so a later phase can wire it into the live loop. A no-document task is graded on one Jx file it writes, so it still cannot score a **project-level** change. A behavior like "asked for a blog, so it turned on the content extension" runs a command tool the runner does not register and leaves nothing for the render critic to mount; it needs a check over `project.json`. Until that exists, those tools are covered by unit tests over the tool surface and the prompt, not by an eval. ## Related - [AI assistant](/docs/studio/ai): what the harness is measuring, from the user's side - [Working in the monorepo](/docs/extending/contributing/monorepo): the testing policy the rest of the repo follows - [Working with agents](/docs/framework/agents): external agents on a Jx project, and the checks that hold them honest ✕ Start here[Overview](/docs/start/)Get started[Install Jx Studio](/docs/start/install/)[Your first project](/docs/start/first-project/)[A tour of Jx Studio](/docs/start/studio-tour/)Tutorials[Tutorial: interactive component](/docs/start/first-component/)[Tutorial: blog with collections](/docs/start/first-collection/)[Video tutorials](/docs/start/videos/)Coming from another tool[Coming from WordPress](/docs/start/coming-from-wordpress/)[Coming from Webflow or Framer](/docs/start/coming-from-webflow/)[Coming from Wix or Squarespace](/docs/start/coming-from-wix/)[Coming from Lovable or Builder.io](/docs/start/coming-from-ai-builders/)Studio[Overview](/docs/studio/)[The desktop app](/docs/studio/desktop/)The workspace[Overview](/docs/studio/interface/)[Welcome screen](/docs/studio/interface/welcome-screen/)[Editors and views](/docs/studio/interface/modes/)[The canvas](/docs/studio/interface/canvas/)[Documents and panes](/docs/studio/interface/tabs/)[Quick Access](/docs/studio/interface/quick-access/)[Keyboard shortcuts](/docs/studio/interface/shortcuts/)[Preferences](/docs/studio/interface/preferences/)[Languages](/docs/studio/interface/languages/)[Problems and progress](/docs/studio/interface/problems-and-progress/)[Commands](/docs/studio/interface/commands/)Projects[Overview](/docs/studio/projects/)[Create a project](/docs/studio/projects/create/)[Pages, layouts, components](/docs/studio/projects/pages-layouts-components/)[The Content workspace](/docs/studio/projects/browse/)[Starter templates](/docs/studio/projects/starters/)[Media](/docs/studio/projects/media/)[Content types](/docs/studio/projects/content-types/)[Project settings](/docs/studio/projects/settings/)[Dependencies and imports](/docs/studio/projects/dependencies/)Edit mode[Overview](/docs/studio/editing/)[Writing and formatting](/docs/studio/editing/writing/)[Slash commands](/docs/studio/editing/slash-commands/)[Frontmatter and metadata](/docs/studio/editing/frontmatter/)[Grid mode](/docs/studio/editing/grid/)Design mode[Overview](/docs/studio/design/)[Outline](/docs/studio/design/layers/)[Insert](/docs/studio/design/elements/)[Content tab](/docs/studio/design/properties/)[Style inspector](/docs/studio/design/style-inspector/)[Breakpoints](/docs/studio/design/breakpoints/)[States and selectors](/docs/studio/design/states-and-selectors/)[Design tokens](/docs/studio/design/tokens/)[Project Styles](/docs/studio/design/stylebook/)[Working with components](/docs/studio/design/components/)[Repeaters](/docs/studio/design/repeaters/)Logic[Overview](/docs/studio/logic/)[Formulas and expressions](/docs/studio/logic/formulas/)[Formula workspace](/docs/studio/logic/formula-workspace/)[Statements](/docs/studio/logic/statements/)[Logic tab](/docs/studio/logic/events/)[Data](/docs/studio/logic/data/)[Data sources](/docs/studio/logic/data-sources/)[Code editing](/docs/studio/logic/code/)Databases[Overview](/docs/studio/data/)[Connections](/docs/studio/data/connections/)[Data tables](/docs/studio/data/tables/)[Data grid](/docs/studio/data/grid/)[Auth and secrets](/docs/studio/data/auth-and-secrets/)AI assistant[Overview](/docs/studio/ai/)[The AI assistant](/docs/studio/ai/chat/)[Document assistant](/docs/studio/ai/document-assistant/)Publish[Overview](/docs/studio/publish/)[Source control](/docs/studio/publish/source-control/)[GitHub](/docs/studio/publish/github/)[Cloudflare Pages](/docs/studio/publish/cloudflare/)[Other hosts](/docs/studio/publish/other-hosts/)[Redirects](/docs/studio/publish/redirects/)[Real-time collaboration](/docs/studio/publish/collaboration/)Framework[Overview](/docs/framework/)Concepts[Documents](/docs/framework/concepts/documents/)[Elements](/docs/framework/concepts/elements/)[Components](/docs/framework/concepts/components/)[Props and scope](/docs/framework/concepts/props-and-scope/)[State](/docs/framework/concepts/state/)[Reactivity](/docs/framework/concepts/reactivity/)[References](/docs/framework/concepts/references/)[Expressions](/docs/framework/concepts/expressions/)[Statements](/docs/framework/concepts/statements/)[Functions and sidecars](/docs/framework/concepts/functions/)[Styling](/docs/framework/concepts/styling/)[Color schemes](/docs/framework/concepts/color-schemes/)[Popovers and overlays](/docs/framework/concepts/overlays/)[Accessibility](/docs/framework/concepts/accessibility/)[Lists and iteration](/docs/framework/concepts/lists/)[Data prototypes](/docs/framework/concepts/data-prototypes/)[Timing](/docs/framework/concepts/timing/)[Dynamic switching](/docs/framework/concepts/switching/)[Security](/docs/framework/concepts/security/)Site architecture[Overview](/docs/framework/site/)[project.json](/docs/framework/site/project-json/)[Routing](/docs/framework/site/routing/)[Layouts](/docs/framework/site/layouts/)[Content collections](/docs/framework/site/content-collections/)[Jx Markdown](/docs/framework/site/jx-markdown/)[Obsidian vault as content](/docs/framework/site/obsidian-vault/)[Relationships](/docs/framework/site/relationships/)[SEO and metadata](/docs/framework/site/seo/)[Locales and languages](/docs/framework/site/i18n/)[Images](/docs/framework/site/images/)[Redirects](/docs/framework/site/redirects/)[Feeds](/docs/framework/site/feeds/)[Site search](/docs/framework/site/search/)[Build output and adapters](/docs/framework/site/deployment/)How compilation works[Overview](/docs/framework/build/)[The dev server](/docs/framework/build/dev-server/)[CLI commands](/docs/framework/build/cli/)Working with agents[Overview](/docs/framework/agents/)[Authoring rules for agents](/docs/framework/agents/authoring-rules/)[Machine-readable docs](/docs/framework/agents/machine-readable/)Reference[Overview](/docs/framework/reference/)[Formula catalog](/docs/framework/reference/formulas/)[Operator reference](/docs/framework/reference/operators/)Extending[Overview](/docs/extending/)Extensions[Extension anatomy](/docs/extending/extensions/anatomy/)[Schema composition](/docs/extending/extensions/schema-composition/)[Custom classes](/docs/extending/extensions/classes/)[Custom formats](/docs/extending/extensions/formats/)[Capability methods](/docs/extending/extensions/capabilities/)[Project sections](/docs/extending/extensions/project-sections/)[Server mounts](/docs/extending/extensions/server/)[Connectors](/docs/extending/extensions/connectors/)[Search indexes](/docs/extending/extensions/search/)[Security and secrets](/docs/extending/extensions/security/)[Tutorial: a TOML format](/docs/extending/extensions/tutorial-toml-format/)[Tutorial: a guestbook](/docs/extending/extensions/tutorial-guestbook/)[First-party extensions](/docs/extending/extensions/first-party/)Embedding[Overview](/docs/extending/embedding/)[Hosting the Studio assets](/docs/extending/embedding/hosting/)[Writing a platform adapter](/docs/extending/embedding/platform-adapter/)[The backend protocol](/docs/extending/embedding/backend-protocol/)[Dev server internals](/docs/extending/embedding/dev-server/)[Embedding the runtime in your app](/docs/extending/embedding/runtime-host/)[Writing assistant tools](/docs/extending/embedding/assistant-harness/)UI kit[The Jx UI kit](/docs/extending/ui-kit/)Reference[Protocol routes](/docs/extending/reference/studio-routes/)[Implementation status](/docs/extending/reference/implementation-status/)[Spec changelog](/docs/extending/reference/spec-changelog/)[Standards alignment](/docs/extending/reference/standards/)Contributing[Working in the monorepo](/docs/extending/contributing/monorepo/)[Contributing to these docs](/docs/extending/contributing/docs/)[AI assistant evals](/docs/extending/contributing/ai-evals/) [![Jx Suite](/favicon.svg) Suite](/)The visual IDE for the web. Build any website, save it as plain files, own it forever. MIT licensed. Product [Studio](/studio)[Download](/download)[Templates](/templates)[Features](/features)[Compare](/compare)[Jx Cloud (soon)](/cloud) Developers [Docs](/docs/start/first-project)[Framework](/framework)[Get Studio](/docs/start/install)[Spec](/docs/framework)[GitHub](https://github.com/jxsuite/jx) Project [MIT License](https://github.com/jxsuite/jx/blob/main/LICENSE)[Releases](https://github.com/jxsuite/jx/releases/latest)[Privacy](/privacy)[Terms](/terms)[Support](/support) Built with Jx Suite. © 2026 Avunu LLC. Open source under the MIT License.