| Filename | Latest commit message | Latest commit date |
|---|---|---|
|
Some checks failed
Run install, format, lint, build, and test on every push / build (push) Has been cancelled
Run install, format, lint, build, and test on every push / publish (push) Has been cancelled
Run install, format, lint, build, and test on every push / publish-github-container-registry (push) Has been cancelled
* Tasks: modern io.modelcontextprotocol/tasks extension (era fork core) Core + test-server implementation of the SEP-2663 tasks extension for modern (2026-07-28) connections, forked from the legacy 2025-11-25 tasks path. SDK v2 removed all tasks support AND era-gates the tasks/* spec methods out of the modern era on both the client (outbound) and server (inbound) sides, and its codec rejects a resultType:"task" result outright. So the Inspector drives the extension itself: task creation rides tools/call (a modern method) with the resultType:"task" frame rewritten at the transport to a CallToolResult carrying the handle in _meta; tasks/get/update/cancel go over a raw-wire request channel that bypasses the era gate (full modern envelope, string ids, consumed by the transport). input_required tasks reuse the existing pending-request modal via fulfilInputRequests, answered with tasks/update. No tasks/list, no tasks/result. Tab + run-as-task gate on the negotiated extension; the managed task store's refresh re-polls known handles on modern (no server list). Test server serves the extension via an Express interceptor (SDK can't) + a shared poll-driven runtime. Full integration coverage of both eras. Refs #1631 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 * Tasks modern era: tests + coverage (raw-wire, schemas, era integration) Adds unit coverage for modernTaskSchemas and the raw-wire request channel (transport-null guard, send failure, timeout, error response, teardown), plus end-to-end era integration tests (create/poll/input-required/update/cancel, synchronous passthrough, unknown-id error, modern store re-poll). Fixes the App and constructor-capabilities tests for the always-on tasks extension. Refs #1631 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 * Tasks modern era: docs + share test servers per era to de-flake coverage Document the tasks era-fork test configs in README and the new core/test-server files in AGENTS.md. Share one modern + one legacy test server across the era integration block (started lazily, stopped in afterAll) — starting a fresh Express server per test made connection negotiation flaky under the heavy concurrent coverage run. Refs #1631 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 * Restore pr-screenshots deleted inadvertently The logging/subscriptions PR screenshots + README were staged-deleted in the working tree before this branch; restore them from v2/main so this PR doesn't remove other PRs' artifacts. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 * Tasks modern era: offer Run-as-task for any tool + PR screenshots On modern (SEP-2663) task creation is server-directed — the client declares the extension once and any tool may return a task — so the Tools screen now offers 'Run as task' for every tool on a modern connection (threaded via a modernTasks prop through InspectorView → ToolsScreen → ToolDetailPanel), rather than gating on the legacy per-tool taskSupport. Verified end-to-end in the web UI (remote proxy): create → poll tasks/get → input_required elicitation → tasks/update → inline result, plus the unchanged legacy tasks/list/result flow. Adds proof screenshots. Refs #1631 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 * Tasks modern era: address two follow-ups (task-input note + ordinary-path auto-poll) 1. The pending-request note now distinguishes a modern task round from an MRTR round: a new 'task-input-required' PendingRequestOrigin threads through fulfilInputRequests → the pending-request modal → MrtrOriginNote, which shows 'submitted via a tasks/update request (SEP-2663), not a retry' instead of the MRTR retry copy. 2. The ordinary (non-run-as-task) modern callTool path now auto-polls an unsolicited task handle: attemptToolCall detects the rewritten handle in the result _meta and drives it to termination (get → input_required/update → inline result), dispatching task events so the Tasks tab + toasts track it and the tool result resolves to the task's final result. The poll loop is shared with the streaming path via extracted helpers (submitModernTaskInput / modernTaskTerminalOutcome / modernTaskPollInterval). Refs #1631 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 * Tasks modern era: address review (negotiated envelope version, input-round cap, clarifying comment) + fresh smoke screenshots Review of #1731: 1. withModernTaskEnvelope now stamps the NEGOTIATED protocol version (getProtocolVersion() ?? MODERN_PROTOCOL_VERSION) so the envelope matches the MCP-Protocol-Version header instead of a hardcoded constant. 2. A never-completing input_required task is now bounded by MRTR_MAX_ROUNDS in the shared submitModernTaskInput (threaded through both poll loops), so a non-conformant server can't re-prompt the user forever. New modern_loop_task test-server tool + integration test cover the cap. 3. ToolDetailPanel: comment clarifying that per-tool taskSupport:"forbidden" is deliberately ignored on modern (task creation is server-directed under SEP-2663). (The abort-during-poll observation is addressed in the PR reply: the cancel path uses tasks/cancel and disconnect rejects the in-flight raw request, so the poll stops; matches legacy.) Fresh web-UI smoke screenshots (modern connected, task completed, input_required with the corrected tasks/update note, input completed). Refs #1631 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 * Tasks modern era: fix uncancelable input-required loop + sticky cancel Two bug fixes reported during manual testing of the tasks era-fork (#1631): 1. A modern task paused at `input_required` (e.g. `modern_loop_task`, which never advances) could not be cancelled: the poll loop blocked on the pending-request modal, so `cancelRequestorTask` sent `tasks/cancel` but the modal kept re-prompting. Now each modern poll loop registers a per-task AbortController (`registerTaskInputAbort`), `cancelRequestorTask` aborts it to reject the pending elicitation and unblock the poll, and the embedded input requests are tagged with their owning task id (`RELATED_TASK_META_KEY`) so the pending-request UI's Cancel action cancels the TASK rather than sending a "cancel" answer the server would re-prompt on. 2. On the legacy server, cancelling a task briefly showed CANCELLED then flipped to COMPLETED when the in-flight tool call resolved. Cancellation is a deliberate terminal decision, so `ManagedRequestorTasksState` now remembers cancelled task ids and ignores any later non-cancelled status update. Also: move the compact task id onto its own line in the embedded monitoring sidebar so the status badge / Cancel Task control don't wrap; DRY the duplicated task poll-interval logic into a shared `taskPollInterval` helper. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 * Refresh tasks PR screenshots + capture both bug-fix scenarios Recapture the five tasks flow screenshots against the current build and add two demonstrating the post-review fixes: modern_loop_task cancelled from the pending-request modal (uncancelable-loop fix) and a legacy task staying CANCELLED after the in-flight call resolves (sticky-cancel fix). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 * Address re-review: sticky cancel on legacy refresh + App cancel-routing test Re-review of the cancelable-loop / sticky-cancel fix raised two non-blocking hardening items: 1. The legacy tasks/list refresh() rebuilt the list from the server, honoring only dismissedTaskIds — so a cooperative-cancel server that re-lists a user-cancelled task as completed could un-stick the cancel on a manual Refresh. The refresh path now pins a cancelledTaskIds task to "cancelled" (keeping it visible-as-cancelled, matching the live-merge handlers) rather than dropping it. Covered by a new state test. 2. Added an App.test.tsx case locking in the user-facing half of the fix: Cancel on a task-input-required elicitation calls cancelRequestorTask (not respond). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XNdjEPKLG637X8YmhDiEk5 --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> |
||
| .. | ||
| .storybook | ||
| public | ||
| server | ||
| src | ||
| static | ||
| .gitignore | ||
| .npmignore | ||
| chromatic.config.json | ||
| eslint.config.js | ||
| index.html | ||
| package-lock.json | ||
| package.json | ||
| README.md | ||
| tsconfig.app.json | ||
| tsconfig.json | ||
| tsconfig.node.json | ||
| tsconfig.storybook.json | ||
| tsconfig.test.json | ||
| tsup.runner.config.ts | ||
| vite.config.ts | ||
| vitest.shims.d.ts | ||
MCP Inspector Web Client
The browser incarnation of the Inspector: a Vite + React + Mantine single-page app backed by a small Node (Hono) server. The SPA is presentational — it renders data and fires callbacks; all MCP state comes from the shared @inspector/core hooks. The backend proxies MCP connections, serves the built SPA, and exposes /api/*.
This README covers what's specific to the web client. For the repo-wide picture (the @inspector/core shared package, the "dumb components" philosophy, the top-level validate/coverage/ci scripts, and publishing), see the root README.
Two halves: src/ (browser) and server/ (Node)
| Path | Runs in | Purpose |
|---|---|---|
src/ |
browser | The React SPA — components, hooks, theme, entry (main.tsx). |
server/ |
Node | The dev/prod backend wiring (never imported by the browser). |
The server/ directory holds the Node-only backend:
vite-hono-plugin.ts— mounts the Hono/api/*middleware onto the Vite dev server (sonpm run devhas a live backend).server.ts— the standalone Hono production server (servesdist/+/api/*).run-web.ts/start-vite-dev-server.ts— entry points the launcher calls for prod--weband--web --dev.web-server-config.ts— env parsing, theGET /api/configpayload, the startup banner.inject-auth-token.ts— embeds the API token into the servedindex.html(see Auth token).sandbox-controller.ts— the MCP Apps sandbox HTTP server;ensure-web-build.ts— buildsdist/on demand for prod--web;vite-base-config.ts— sharedoptimizeDepsexclusions.
Development
npm run dev # Vite dev server + Hono /api middleware, HMR
For the launcher-driven prod/dev flows (npm run web / web:dev), see the root README — those run the built launcher.
Build
npm run build # tsc -b → vite build → build:runner
Two artifacts come out, both of which ship in the published package:
dist/— the browser SPA (vite build). Served statically by the prod backend.build/— the Node prod-server runner (build:runner=tsup --config tsup.runner.config.ts), which bundlesserver/+@inspector/coreinto one ESM file and externalizes npm deps.
build:client runs only the vite build half when you just need fresh dist/.
Component layers
Components live under src/components/ in four layers, smallest to largest:
| Layer | Count | What it is |
|---|---|---|
elements/ |
~31 | Leaf presentational pieces (badges, buttons, toggles) over Mantine primitives. |
groups/ |
~63 | Composite pieces (cards, panels, modals, control bars). |
screens/ |
~11 | Full tab screens (Tools, Resources, Servers, monitoring screens…). |
views/ |
1 | InspectorView — the top-level layout that composes the screens. |
Every screen and element has a *.stories.tsx (see Storybook). Styling follows the Mantine-first rules in AGENTS.md — theme variants and component props over CSS, --inspector-* tokens over raw colors.
MCP Apps screen automation contract
The Apps screen exposes a small, stable set of data-testid / data-* attributes so an automated driver (deep-link auto-open, CI review harness) can waitForSelector on a deterministic signal instead of sleeping. Treat these as a public contract — drivers depend on them staying stable:
| Attribute | Where | Meaning |
|---|---|---|
data-testid="apps-form" |
Apps content card | The container that carries the status/error attributes below. |
data-app-status |
on apps-form |
Renderer lifecycle: idle (nothing running) → loading (bridge building / ui/initialize in flight) → ready (view fired notifications/initialized) → error (bridge factory threw/rejected). Poll for ready. |
data-app-error |
on apps-form |
The failure reason string when data-app-status="error" (e.g. no connected client); absent otherwise. |
data-testid="apps-error" |
error panel | Rendered below the frame when the app fails to load (factory throw/reject); shows the reason so the failure isn't a silent blank frame. |
data-testid="open-app" |
Open App button | Launches the selected app. |
data-testid="apps-stage" |
Stage-partial button | Snapshots the current form values for progressive-render testing. |
data-testid="apps-messages" |
messages panel | ui/message submissions from the running view. |
data-testid="apps-logs" |
app-logs panel | notifications/message log entries (default-expanded). |
The renderer lifecycle itself is AppRendererStatus (loading | ready | error) reported via AppRenderer's onAppStatusChange; the screen maps it to data-app-status. Resource-read failures (malformed/404 UI resource) are surfaced as a toast via the bridge factory's onResourceError; because the app never reaches ready in that case, a driver times out on data-app-status and reads the toast.
Deep-link auto-connect
A driver (launcher, CLI --print-handoff, CI review harness) can reach a connected inspector with a single navigate by encoding the target in the URL query string. Parsing + security gating live in src/utils/deepLink.ts (parseDeepLink), and a returned DeepLink is proof the link passed validation.
http://127.0.0.1:6274/?serverUrl=<url>&transport=http|sse&autoConnect=<token>
| Param | Meaning |
|---|---|
serverUrl |
The MCP server URL. Restricted to http: / https: (a crafted javascript: / data: / file: value is rejected). Canonicalized via URL.href so it matches the OAuth store's key form. |
transport |
http (streamable-HTTP, the default) or sse. Unknown values fall back to http. |
autoConnect |
CSRF gate. Must equal the per-launch MCP_INSPECTOR_API_TOKEN. The token is random per launch and only known to whatever started the server, so a third-party-minted link cannot satisfy it — this is the same exposure surface as the existing ?MCP_INSPECTOR_API_TOKEN= param. Without a match the link is ignored. |
The deep link upserts a stable deep-link catalog row (so a reload reconnects to the same row instead of accumulating duplicates) and connects. Connection-level outcomes are surfaced on the AppShell.Header as a machine-readable contract, so a driver can waitForSelector and read the failure reason without scraping a transient toast:
| Attribute | Where | Meaning |
|---|---|---|
data-testid="connection-status" |
header | The element carrying the attributes below. |
data-status |
on connection-status |
The live ConnectionStatus (disconnected → connecting → connected / error). Poll for connected. |
data-error-message |
on connection-status |
Why the last connect failed (handshake error, OAuth-start failure, deep-link automation failure); absent when there is no error. |
data-deeplink |
on connection-status |
parsed (a valid deep link drove this load), rejected (deep-link params present but the token/serverUrl gate failed), or none. Lets a driver distinguish "no deep link" from "rejected" — both otherwise leave data-status idle. |
Landing on a rendered app
Three further params extend the deep link to pre-select — and optionally auto-open — an MCP App, so a driver reaches a rendered widget with zero clicks:
…&openApp=<toolName>&appArgs=<base64url(JSON)>&autoOpen=<token>
| Param | Meaning |
|---|---|
openApp |
The app-tool name. Once the connection is up and the tool appears in the app list, the inspector switches to the Apps tab and pre-selects it. |
appArgs |
base64url(JSON) object of form values. Merged over the tool's schema defaults (collectSchemaDefaults) so a required-with-default field isn't left blank — which would otherwise disable "Open App". Malformed / non-object values fall back to {}. |
autoOpen |
Same CSRF gate as autoConnect — must equal the session token. When set, "Open App" fires automatically (a tool call from a URL), so the token gate is mandatory. Without a match the app is pre-selected but not opened. |
The app-side render lifecycle is observable through the MCP Apps screen automation contract above (data-app-status="ready"), so a driver can waitForSelector the whole connect → open → ready chain deterministically.
Theme (src/theme/)
Each customized Mantine component has a Theme<Name>.ts file (Button.ts, Text.ts, …, ~21 total) exporting a Theme<Name> constant; the barrel index.ts re-exports them and theme.ts assembles the MantineProvider theme. Theme files hold app-wide defaults and variants (flat CSS-in-JS); only pseudo-selectors, nested child selectors, keyframes, and native-HTML styling belong in App.css. Element components import from @mantine/core (never from theme/) — the theme layer is applied transparently by the provider.
Testing
Tests run under three Vitest projects (configured in vite.config.ts), each in the right environment:
| Project | Env | Scope | Script |
|---|---|---|---|
unit |
happy-dom | Components, hooks, utils (*.test.tsx beside the source) |
npm test |
integration |
node | @inspector/core + transports + auth, spawning the real stdio test server (src/test/integration/**) |
npm run test:integration |
storybook |
real Chromium | Story play functions as interaction tests | npm run test:storybook |
npm testruns the fast unit project (happy-dom).test:watchfor the loop.- Integration tests run in a real Node env (no happy-dom, 30s timeouts) and spawn
test-servers/build/test-server-stdio.jsas a subprocess, sopretest/the coverage script build the test servers first (test-servers:build). npm run test:coverageruns unit and integration under v8 instrumentation and enforces the per-file ≥90% gate (lines/statements/functions/branches) — the same gate CI runs. Genuinely-unreachable branches are annotated with a justified/* v8 ignore … */, not waved through.
Integration tests live under src/test/integration/ mirroring the core/ layout; anything placed there is picked up by the integration project automatically. Render components with renderWithMantine (src/test/renderWithMantine.tsx) so they get the project theme.
Storybook
npm run storybook # dev server on :6006
npm run build:storybook # static build
npm run test:storybook # run every story's play function in headless Chromium
Storybook is first-class here because the components are presentational — each renders against fixture props. Play functions double as interaction tests and run headless in real Chromium via @vitest/browser-playwright + @storybook/addon-vitest (the storybook Vitest project). They're part of npm run ci (which installs the Chromium binary first) but kept out of the fast validate loop since they need the browser.
Auth token
The dev/prod backend guards every /api/* route with x-mcp-remote-auth: Bearer <MCP_INSPECTOR_API_TOKEN>. The browser recovers the token, in priority order (see App.tsx getAuthToken()): the window.__INSPECTOR_API_TOKEN__ global injected into index.html on every page load (server/inject-auth-token.ts), then a ?MCP_INSPECTOR_API_TOKEN=… query param, then sessionStorage. Injection is a no-op when auth is disabled (DANGEROUSLY_OMIT_AUTH). See the root AGENTS.md for the full rationale.
HTTP proxy support
The web backend connects to remote MCP servers through the shared Node transport (core/mcp/node/transport.ts), which honors the conventional proxy environment variables: HTTPS_PROXY / HTTP_PROXY (and their lowercase forms) select the proxy, and NO_PROXY exempts hosts. Routing is powered by undici's EnvHttpProxyAgent, imported lazily only when a proxy variable is set, so runs without a proxy configured pay no cost. See the CLI README for more detail.