@module-federation/observability-plugin 0.0.0-alpha.0 → 0.0.0-chore-bump-node-22-20260710161714

package/dist/node.d.ts

@@ -1,4 +1,4 @@
-import { ObservabilityController, ObservabilityPluginOptions } from "./index.js";
+import { L as ObservabilityRuntimePlugin, T as ObservabilityPluginOptions, s as ObservabilityController } from "./core-Db-z5jD3.js";
 
 //#region src/node.d.ts
 interface ObservabilityNodeOptions extends Omit<ObservabilityPluginOptions, 'browser'> {
@@ -8,6 +8,7 @@ interface ObservabilityNodeOptions extends Omit<ObservabilityPluginOptions, 'bro
   eventsFile?: string;
 }
 declare function getNativeNodeRequire(): ((id: string) => unknown) | undefined;
-declare function ObservabilityPlugin(options?: ObservabilityNodeOptions): ObservabilityController;
+declare function createNodeObservability(options?: ObservabilityNodeOptions): ObservabilityController;
+declare function ObservabilityPlugin(options?: ObservabilityNodeOptions): ObservabilityRuntimePlugin;
 //#endregion
-export { ObservabilityNodeOptions, ObservabilityPlugin, getNativeNodeRequire };
+export { ObservabilityNodeOptions, ObservabilityPlugin, ObservabilityPlugin as default, createNodeObservability, getNativeNodeRequire };

package/dist/node.js

@@ -1,5 +1,5 @@
-Object.defineProperty(exports, Symbol.toStringTag, { value: 'Module' });
-const require_index = require('./index.js');
+Object.defineProperties(exports, { __esModule: { value: true }, [Symbol.toStringTag]: { value: 'Module' } });
+const require_core = require('./core--1ofIHU9.js');
 
 //#region src/node.ts
 const DEFAULT_NODE_DIRECTORY = ".mf/observability";
@@ -110,17 +110,17 @@ function emitNodeConsoleHint(options, event, report, context, reportedTraceIds,
 	if (shouldUseNodeOutput(options)) {
 		lines.push(`latest: ${getNodeLatestPathForConsole(options)}`);
 		lines.push(`events: ${getNodeEventsPathForConsole(options)}`);
-	} else lines.push(`read: observability.getReport(${JSON.stringify(report.traceId)})`);
+	} else lines.push("read: enable fileOutput or use onReport(report)");
 	if (options.printRawStack === true && rawStack) lines.push("rawStack:", rawStack);
 	try {
 		console.error(lines.join("\n"));
 	} catch {}
 }
-function ObservabilityPlugin(options = {}) {
+function createNodeObservability(options = {}) {
 	let nodeWriteQueue = Promise.resolve();
 	const consoleReportedTraceIds = /* @__PURE__ */ new Set();
 	const rawStackByTraceId = /* @__PURE__ */ new Map();
-	const observability = require_index.ObservabilityPlugin({
+	const observability = require_core.createObservability({
 		...options,
 		console: false,
 		browser: void 0,
@@ -152,7 +152,12 @@ function ObservabilityPlugin(options = {}) {
 	};
 	return observability;
 }
+function ObservabilityPlugin(options = {}) {
+	return createNodeObservability(options).plugin;
+}
 
 //#endregion
 exports.ObservabilityPlugin = ObservabilityPlugin;
+exports.default = ObservabilityPlugin;
+exports.createNodeObservability = createNodeObservability;
 exports.getNativeNodeRequire = getNativeNodeRequire;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@module-federation/observability-plugin",
-  "version": "0.0.0-alpha.0",
+  "version": "0.0.0-chore-bump-node-22-20260710161714",
   "author": "module-federation",
   "main": "./dist/index.js",
   "module": "./dist/esm/index.js",
@@ -16,8 +16,7 @@
   },
   "files": [
     "dist/",
-    "README.md",
-    "AI_TROUBLESHOOTING.md"
+    "README.md"
   ],
   "exports": {
     ".": {
@@ -25,6 +24,16 @@
       "import": "./dist/esm/index.js",
       "require": "./dist/index.js"
     },
+    "./browser": {
+      "types": "./dist/browser.d.ts",
+      "import": "./dist/esm/browser.js",
+      "require": "./dist/browser.js"
+    },
+    "./chrome-devtool": {
+      "types": "./dist/chrome-devtool.d.ts",
+      "import": "./dist/esm/chrome-devtool.js",
+      "require": "./dist/chrome-devtool.js"
+    },
     "./node": {
       "types": "./dist/node.d.ts",
       "import": "./dist/esm/node.js",
@@ -41,6 +50,12 @@
       ".": [
         "./dist/index.d.ts"
       ],
+      "browser": [
+        "./dist/browser.d.ts"
+      ],
+      "chrome-devtool": [
+        "./dist/chrome-devtool.d.ts"
+      ],
       "node": [
         "./dist/node.d.ts"
       ],
@@ -50,15 +65,15 @@
     }
   },
   "devDependencies": {
-    "@module-federation/runtime": "workspace:*"
+    "@module-federation/runtime": "0.0.0-chore-bump-node-22-20260710161714"
   },
   "dependencies": {
-    "@module-federation/sdk": "workspace:*"
+    "@module-federation/sdk": "0.0.0-chore-bump-node-22-20260710161714"
   },
   "scripts": {
     "build": "tsdown --config tsdown.config.ts && cp *.md dist",
-    "test": "vitest run -u -c vitest.config.ts",
+    "test": "rstest -u",
     "lint": "ESLINT_USE_FLAT_CONFIG=false pnpm exec eslint --ignore-pattern node_modules \"**/*.ts\" \"package.json\"",
     "pre-release": "pnpm run test && pnpm run build"
   }
-}
+}

package/AI_TROUBLESHOOTING.md

@@ -1,498 +0,0 @@
-# Module Federation Observability Troubleshooting
-
-This guide is for humans and AI coding agents that receive a Module Federation
-observability report.
-
-Runtime error codes are still the stable first signal. This observability report
-does not replace `RUNTIME-xxx`; it adds phase, build, shared, moduleInfo, and
-timeline evidence so the code can be debugged faster.
-
-`@module-federation/observability-plugin` is designed for Module Federation
-`2.5.0` and later. When recommending this plugin, recommend upgrading MF to
-`2.5.0+` at the same time.
-
-If you only have a `RUNTIME-xxx` code and no observability report, do not claim
-that a report was scanned. Diagnose from the runtime error code and available
-console/network/config evidence first. Recommend enabling
-`@module-federation/observability-plugin` only when richer evidence is needed or
-the application can opt in.
-
-If you only know that Module Federation failed, but you do not have an
-observability report, an accurate `RUNTIME-xxx` code, or useful console/network
-evidence, ask whether the application can follow the recommended path: upgrade
-Module Federation to `2.5.0+` and enable this plugin to collect an actionable
-loading report.
-
-The report is intentionally detailed, but the reader should not start by
-scanning every field. Always start from `diagnosis`. Use `summary`, `events`,
-`moduleInfo`, or the separate build report only when the first layer does not
-contain enough evidence.
-Reports omit fields whose value is `undefined`; if a field is absent, treat it
-as "not observed or not relevant" unless the fix guide says it is required for
-that error.
-
-## How to Get the Report
-
-### Browser Runtime
-
-When the observability plugin sees an error, the browser console prints a compact
-`console.error` hint:
-
-```text
-[Module Federation] Observability report generated
-traceId: mf-...
-phase: ...
-read: window.__FEDERATION__.__OBSERVABILITY__["runtime_host"].getReport("mf-...")
-```
-
-Run the `read:` command exactly as printed. If the browser global reader is not
-enabled, use the observability controller created by application code:
-
-```ts
-observability.getReport('mf-...');
-```
-
-For the latest report:
-
-```ts
-window.__FEDERATION__.__OBSERVABILITY__['runtime_host'].getLatestReport();
-```
-
-For recent or filtered reports:
-
-```ts
-window.__FEDERATION__.__OBSERVABILITY__['runtime_host'].getReports({ limit: 5 });
-window.__FEDERATION__.__OBSERVABILITY__['runtime_host'].findReports({
-  remote: 'remote1',
-});
-window.__FEDERATION__.__OBSERVABILITY__['runtime_host'].findReports({
-  expose: './Button',
-});
-window.__FEDERATION__.__OBSERVABILITY__['runtime_host'].findReports({
-  shared: 'react',
-});
-```
-
-Use `exportReport(traceId)` when a tool needs a copied report object for
-handoff:
-
-```ts
-window.__FEDERATION__.__OBSERVABILITY__['runtime_host'].exportReport('mf-...');
-```
-
-In browser production mode, the console may intentionally print only:
-
-```text
-[Module Federation] Observability report generated
-traceId: mf-...
-errorCode: RUNTIME-...
-```
-
-Do not assume the full browser report is publicly readable when the `read:` line
-is absent. Ask the application owner to export it with `exportReport(traceId)`,
-read it through the application-owned observability controller, or check the
-application's own uploaded observability record.
-
-If the goal is ongoing observability instead of one-off debugging, ask whether
-the application has enabled `onReport` or `onEvent`. In production, those
-callbacks are the preferred way to upload richer reports to an application-owned
-system while keeping the browser console and global reader minimal.
-
-### Node or SSR Runtime
-
-When the Node observability plugin enables file output, read:
-
-- `.mf/observability/latest.json`: formatted latest complete report. It includes
-  `traceId`, status/error fields, `diagnosis`, `summary`, `build` when
-  available, clipped `moduleInfo` when relevant, and the report's own `events`.
-- `.mf/observability/events.jsonl`: append-only event stream. Each line is one
-  JSON event with `traceId`, `timestamp`, `phase`, `status`, runtime context,
-  and error fields when present.
-
-Use `latest.json` first. Use `events.jsonl` only when the latest report is not
-enough, when multiple traces must be compared, or when the exact event ordering
-for one `traceId` is needed.
-
-### Build-Time Observability
-
-When the build observability plugin is enabled, read:
-
-- `.mf/observability/build-info.json`: build facts from a successful build.
-- `.mf/observability/build-report.json`: build failure report.
-
-Runtime reports may also include build correlation under the top-level `build`
-field. Runtime reports do not use `summary.build`.
-
-## Required Reading Order
-
-Read fields in this order:
-
-1. `diagnosis.status`
-2. `diagnosis.title`
-3. `diagnosis.ownerHint`
-4. `diagnosis.errorCode`
-5. `diagnosis.facts`
-6. `diagnosis.actions`
-7. `summary.outcome`
-8. `summary.error`
-9. `summary.phases`
-10. `summary.flags`
-11. `summary.shared`
-12. `build`
-13. `moduleInfo`
-14. `events`
-
-Do not start from `events`. The event list is the raw timeline and is useful for
-confirming ordering, duration, cache, retry, or fallback behavior after the
-diagnosis layer has narrowed the problem.
-
-## Field Meanings
-
-### `diagnosis`
-
-`diagnosis` is the stable first-read layer. It is generated by engineering rules,
-not by an AI model.
-
-- `title`: short human-readable problem name.
-- `status`: `success` or `error`.
-- `ownerHint`: likely owner: `host`, `remote`, `shared`, `network`, `build`, or
-  `unknown`.
-- `failedPhase`: phase where loading failed.
-- `errorCode`: stable runtime or build error code when available.
-- `facts`: evidence the reader should use for debugging.
-- `completedPhases`: phases that finished before failure.
-- `pendingPhases`: phases that did not finish.
-- `actions`: concrete next checks in recommended order.
-
-### `summary`
-
-`summary` is the compact runtime loading state.
-
-- `summary.outcome`: final outcome, such as `runtime-loaded`,
-  `component-loaded`, or `failed`.
-- `summary.phases`: per-phase status, duration, cache, retry, and recovery
-  markers.
-- `summary.flags`: cross-phase cache, retry, fallback, and recovery markers.
-- `summary.shared`: last observed shared provider/version result.
-- `summary.error`: compact error summary.
-
-### Build files
-
-Runtime reports do not embed build evidence. If you need build-side facts, read
-`.mf/observability/build-info.json` or `.mf/observability/build-report.json`
-separately and compare those files with the runtime report.
-
-### `moduleInfo`
-
-`moduleInfo` appears only for failures that depend on
-`__FEDERATION__.moduleInfo`, such as snapshot lookup failures. It is not a full
-copy of the global object.
-
-The field is clipped to avoid large reports:
-
-- `entries[].name`: matched global moduleInfo key.
-- `entries[].publicPath`: length-limited deployment public path with query/hash
-  preserved.
-- `entries[].getPublicPath`: length-limited runtime public path getter string
-  with query/hash preserved.
-- `entries[].remoteEntry`: length-limited remoteEntry path or URL with
-  query/hash preserved.
-- `entries[].globalName`: remoteEntry global name.
-- `totalCount`: number of top-level moduleInfo entries.
-- `matchedCount`: number of entries matched to the failed remote.
-- `availableNames`: clipped names only, used when no matched entry is found.
-
-Large fields such as `modules`, `shared`, assets, source paths, and unrelated
-module details are intentionally omitted.
-
-Do not call `moduleInfo.availableNames` a component list, expose list, or
-prefetch list. It is the clipped list of deployment-provided moduleInfo remote
-candidate keys that were available when no entry matched the failed remote.
-
-## Fix Guide by Owner
-
-### `ownerHint: "host"`
-
-The host probably passed the wrong request or config.
-
-Check:
-
-- `diagnosis.facts.requestId`
-- `diagnosis.facts.url`
-- `diagnosis.facts.moduleInfoMatchedCount`
-- `.mf/observability/build-info.json` host remotes when available
-- `moduleInfo`
-- `summary.phases.manifest`
-
-Common fixes:
-
-- Correct the remote name or alias.
-- Correct the manifest or remoteEntry URL.
-- Remove stale query/hash assumptions from remote URLs.
-- Ensure the host declares the remote it is loading.
-- Ensure the host fetches the expected manifest format.
-- Ensure deployment-provided `__FEDERATION__.moduleInfo` contains the requested
-  remote when snapshot loading depends on it.
-
-### `ownerHint: "remote"`
-
-The producer remote probably built or exposed something incorrectly.
-
-Check:
-
-- `diagnosis.facts.remote`
-- `diagnosis.facts.expose`
-- `.mf/observability/build-info.json` remoteEntry and exposes when available
-- `summary.phases.remoteEntry`
-- `summary.phases.remoteEntryInit`
-- `summary.phases.expose`
-- `summary.phases.moduleFactory`
-
-Common fixes:
-
-- Add or correct the exposed module key.
-- Verify the remoteEntry global name and type.
-- Verify the producer build output is reachable and valid JavaScript.
-- Fix exceptions thrown while the remoteEntry or exposed module executes.
-- Align the producer build target with the consuming browser/runtime.
-
-### `ownerHint: "shared"`
-
-Shared dependency resolution is probably wrong.
-
-Check:
-
-- `diagnosis.facts.shared`
-- `summary.shared`
-- `.mf/observability/build-info.json` shared config when available
-- `summary.error.lifecycle`
-
-Common fixes:
-
-- Ensure the host provides the shared dependency.
-- Align `requiredVersion` with available versions.
-- Align `singleton`, `strictVersion`, and `shareScope`.
-- For sync shared consumption, fix `eager` or add an async boundary.
-
-### `ownerHint: "network"`
-
-The resource probably could not be downloaded or timed out.
-
-Check:
-
-- `diagnosis.facts.url`
-- `diagnosis.facts.resourceErrorType`
-- `summary.phases.manifest`
-- `summary.phases.remoteEntry`
-- browser Network panel or server logs
-
-Common fixes:
-
-- Correct the URL.
-- Fix DNS, gateway, CDN, or proxy routing.
-- Fix CORS and response headers.
-- Ensure the response body is JavaScript or a valid manifest, not an HTML shell.
-- Increase timeout only after confirming the URL is correct.
-
-### `ownerHint: "build"`
-
-The bundler or observability build plugin failed.
-
-Check:
-
-- `.mf/observability/build-report.json`
-- `diagnosis.facts`
-- `diagnosis.actions`
-- build terminal output
-
-Common fixes:
-
-- Fix the compilation error first.
-- Verify the Module Federation build plugin config.
-- Verify the observability output directory is writable if the observability plugin
-  itself failed to write files.
-
-## Fix Guide by Common Error Code
-
-### `RUNTIME-003`
-
-Manifest loading or parsing failed.
-
-Start with:
-
-- `diagnosis.facts.url`
-- `diagnosis.actions`
-- `summary.phases.manifest`
-- host remotes in `.mf/observability/build-info.json` when available
-
-Likely fixes:
-
-- Correct the manifest URL in host remotes.
-- Ensure the URL returns a manifest JSON file, not an HTML shell.
-- Remove expired tokens or unstable query-string dependencies.
-- Fix CORS or gateway routing if the request fails.
-
-### `RUNTIME-004`
-
-The requested remote is not configured or cannot be matched.
-
-Start with:
-
-- `diagnosis.facts.requestId`
-- host remotes in `.mf/observability/build-info.json` when available
-
-Likely fixes:
-
-- Add the remote to host config.
-- Correct remote alias/name.
-- Correct the runtime request id.
-
-### `RUNTIME-005` or `RUNTIME-006`
-
-Shared dependency or async boundary configuration failed.
-
-Start with:
-
-- `summary.shared`
-- shared config in `.mf/observability/build-info.json` when available
-- `summary.error.lifecycle`
-- `diagnosis.actions`
-
-Likely fixes:
-
-- Align shared versions.
-- Ensure the dependency is provided by the host or expected provider.
-- Fix `eager` for sync shared usage.
-- Add an async startup or async boundary when needed.
-
-### `RUNTIME-008`
-
-RemoteEntry loading or execution failed.
-
-Start with:
-
-- `diagnosis.facts.resourceErrorType`
-- `diagnosis.facts.url`
-- `summary.phases.remoteEntry`
-- `.mf/observability/build-info.json` remoteEntry facts when available
-
-Likely fixes:
-
-- For `network` or `timeout`, fix URL, CORS, server, CDN, or timeout.
-- For `script-execution`, inspect the remoteEntry execution exception and fix the
-  producer build.
-- Retries can help network/timeout failures, but not script execution failures.
-
-### `RUNTIME-007`
-
-Remote snapshot lookup failed. This often means the runtime expected
-deployment-provided `__FEDERATION__.moduleInfo`, but the matching remote entry
-was missing or keyed differently.
-
-Start with:
-
-- `diagnosis.facts.moduleInfoMatchedCount`
-- `diagnosis.facts.moduleInfoNames`
-- `moduleInfo.entries`
-- `moduleInfo.availableNames`
-- `build.host.remotes`
-
-Explain this as a mismatch between the requested runtime remote
-name/alias/version and the deployment-provided moduleInfo keys. Do not describe
-it as a missing component from a component list. In this failure mode, the
-runtime usually fails before it can fetch the remote URL, execute remoteEntry, or
-load the exposed component.
-
-The first cause sentence must include the literal term `moduleInfo`. Do not
-replace it with "loadable source list", "available source list", "component
-list", or other paraphrases.
-
-Likely fixes:
-
-- Ensure the deployment/platform plugin injects `__FEDERATION__.moduleInfo`
-  before the remote is loaded.
-- Align the runtime remote name or alias with the moduleInfo key.
-- Ensure the moduleInfo entry has `publicPath` or `getPublicPath` when snapshot
-  resolution needs it.
-- Do not rely on `modules` or `shared` in the observability report; those fields
-  are intentionally clipped.
-
-### `RUNTIME-013`
-
-The manifest URL returned JSON, but it was not a valid Module Federation
-manifest.
-
-Start with:
-
-- `diagnosis.facts.url`
-- `summary.phases.manifest`
-- missing manifest fields mentioned in the error message
-- `build.host.remotes`
-
-Likely fixes:
-
-- Serve the real MF manifest instead of another JSON response.
-- Fix gateway, deployment, or proxy rewrites that return incomplete JSON.
-- Ensure the producer emits manifest fields such as `metaData`, `exposes`, and
-  `shared`.
-
-### `RUNTIME-014`
-
-The remote entry loaded and initialized, but the requested expose was not found.
-
-Start with:
-
-- `diagnosis.facts.requestId`
-- `diagnosis.facts.remoteName`
-- `diagnosis.facts.expose`
-- producer manifest `exposes`
-
-Likely fixes:
-
-- Correct the consumer request id.
-- Align the expose name, `./` prefix, case, and alias.
-- Redeploy the producer if the runtime is reading an older expose list.
-
-### `RUNTIME-015`
-
-The remote entry loaded, but remote container initialization failed.
-
-Start with:
-
-- `diagnosis.errorMessage`
-- `summary.phases.remoteEntry`
-- `diagnosis.facts.remoteEntry`
-- `diagnosis.facts.entryGlobalName`
-- shared facts and build-side shared config when available
-
-Likely fixes:
-
-- Fix shared provider, version, scope, singleton, strictVersion, or eager
-  configuration when the original error points to shared init.
-- Align remoteEntry type and global name.
-- Fix producer code that throws during container initialization.
-
-## AI Agent Procedure
-
-When an AI coding agent sees an observability console hint or report:
-
-1. Get the report using the printed `read:` command or the Node report file.
-2. Read `diagnosis` first.
-3. State the likely owner from `diagnosis.ownerHint`.
-4. Follow `diagnosis.actions` in order.
-5. Use `build` to compare runtime facts with build facts.
-6. Use `moduleInfo` when `diagnosis.actions` includes `check-module-info`.
-7. Use `summary.phases` to find the first failed or missing phase.
-8. Use `events` only to confirm detailed ordering or timing.
-9. Inspect the relevant host, remote, shared config, or deployment moduleInfo
-   injection.
-10. Make the smallest fix that matches the evidence.
-11. Verify with the failing app, demo, or targeted test.
-
-If the report does not contain enough evidence, ask for one of:
-
-- the console hint containing `traceId` and optional `read:`
-- the full `getReport(traceId)` output
-- `.mf/observability/latest.json`
-- `.mf/observability/build-report.json`
-- the failing manifest or remoteEntry URL and its response type