Merge branch 'develop' into feat/footnotes

Resolve conflicts at shared registration points by unioning both features
(footnotes + the already-merged html-embed / page-embed work):
- slash-menu/menu-items.ts, editor extensions.ts: keep both imports + configures
- collaboration.util.ts: register footnote nodes and pageEmbed
- editor-ext marked.utils.ts: register footnote + html-embed markdown extensions
- editor-ext package.json/tsconfig.json/vitest.config.ts: union of test config
  (jsdom env for footnote DOM tests + combined test/spec include glob)

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
claude_code
2026-06-20 22:21:07 +03:00
193 changed files with 14325 additions and 3395 deletions

View File

@@ -5,7 +5,8 @@
"scripts": {
"build": "tsc --build",
"dev": "tsc --watch",
"test": "vitest run"
"test": "vitest run",
"test:watch": "vitest"
},
"main": "dist/index.js",
"module": "./src/index.ts",

View File

@@ -16,12 +16,14 @@ export * from "./lib/custom-code-block";
export * from "./lib/drawio";
export * from "./lib/excalidraw";
export * from "./lib/embed";
export * from "./lib/html-embed/html-embed";
export * from "./lib/mention";
export * from "./lib/markdown";
export * from "./lib/search-and-replace";
export * from "./lib/embed-provider";
export * from "./lib/subpages";
export * from "./lib/transclusion";
export * from "./lib/page-embed";
export * from "./lib/highlight";
export * from "./lib/indent";
export * from "./lib/heading/heading";

View File

@@ -0,0 +1,138 @@
import { Node, mergeAttributes } from "@tiptap/core";
import { ReactNodeViewRenderer } from "@tiptap/react";
export interface HtmlEmbedOptions {
HTMLAttributes: Record<string, any>;
view: any;
}
export interface HtmlEmbedAttributes {
// Raw HTML/CSS/JS string that is injected verbatim into the wiki origin.
source?: string;
}
declare module "@tiptap/core" {
interface Commands<ReturnType> {
htmlEmbed: {
setHtmlEmbed: (attributes?: HtmlEmbedAttributes) => ReturnType;
};
}
}
/**
* Encode the raw source to base64 for the `data-source` attribute.
*
* The source is arbitrary HTML/CSS/JS. Storing it raw inside an HTML attribute
* would (a) require heavy escaping and (b) risk the parser interpreting markup
* inside the attribute. Base64 makes the round-trip HTML <-> ProseMirror JSON
* lossless and keeps the markup inert while it sits in the attribute.
*
* `encodeURIComponent`/`decodeURIComponent` wrap btoa/atob so that non-Latin1
* (UTF-8) characters survive the base64 step.
*/
export function encodeHtmlEmbedSource(source: string): string {
if (!source) return "";
try {
if (typeof btoa === "function") {
return btoa(encodeURIComponent(source));
}
// Node fallback (server-side schema parsing has no global btoa).
return Buffer.from(encodeURIComponent(source), "utf-8").toString("base64");
} catch {
// Never swallow silently in a way that loses data: fall back to raw.
return "";
}
}
export function decodeHtmlEmbedSource(encoded: string): string {
if (!encoded) return "";
try {
if (typeof atob === "function") {
return decodeURIComponent(atob(encoded));
}
// Node fallback.
return decodeURIComponent(
Buffer.from(encoded, "base64").toString("utf-8"),
);
} catch {
return "";
}
}
export const HtmlEmbed = Node.create<HtmlEmbedOptions>({
name: "htmlEmbed",
inline: false,
group: "block",
// atom + isolating: the node has no editable ProseMirror children; its body
// is the opaque `source` string rendered by the NodeView.
atom: true,
isolating: true,
defining: true,
draggable: true,
addOptions() {
return {
HTMLAttributes: {},
view: null,
};
},
addAttributes() {
return {
source: {
default: "",
// Decode the base64 payload back to the raw source on parse.
parseHTML: (element) =>
decodeHtmlEmbedSource(element.getAttribute("data-source") || ""),
// Encode the raw source to base64 on render so it round-trips losslessly
// through the HTML <-> JSON conversions used by export/import/collab.
renderHTML: (attributes: HtmlEmbedAttributes) => ({
"data-source": encodeHtmlEmbedSource(attributes.source || ""),
}),
},
};
},
parseHTML() {
return [
{
tag: `div[data-type="${this.name}"]`,
},
];
},
renderHTML({ HTMLAttributes }) {
// The static HTML representation is just a marker div carrying the encoded
// source. The actual raw markup is NOT expanded here on purpose: the static
// generateHTML output (used for previews, search indexing, exports) must not
// itself become an injection vector. Only the client NodeView expands and
// executes the source.
return [
"div",
mergeAttributes(
{ "data-type": this.name },
this.options.HTMLAttributes,
HTMLAttributes,
),
];
},
addCommands() {
return {
setHtmlEmbed:
(attrs: HtmlEmbedAttributes) =>
({ commands }) => {
return commands.insertContent({
type: this.name,
attrs: attrs,
});
},
};
},
addNodeView() {
// Force the react node view to render immediately using flush sync.
this.editor.isInitialized = true;
return ReactNodeViewRenderer(this.options.view);
},
});

View File

@@ -0,0 +1,112 @@
import { describe, it, expect } from "vitest";
import { markdownToHtml, htmlToMarkdown } from "./index";
import {
encodeHtmlEmbedSource,
decodeHtmlEmbedSource,
} from "../html-embed/html-embed";
// SECURITY (Variant C admin gate, import attack surface).
//
// The markdown import path is the only write path where an htmlEmbed reaches
// the server purely from file bytes (no editor / collab socket). The marked
// tokenizer in `html-embed.marked.ts` and the turndown rule in
// `turndown.utils.ts` are what materialize the `<!--html-embed:BASE64-->`
// marker into the `<div data-type="htmlEmbed" data-source="BASE64">` element
// that the server then parses into an htmlEmbed node and the admin gate strips.
//
// If either the tokenizer regex or the turndown rule shape drifts, the marker
// would either (a) stop becoming an htmlEmbed node (silently dropping admin
// content) or (b) become some OTHER tag the server's `hasHtmlEmbedNode` no
// longer recognizes (a strip bypass). These tests pin the marker <-> embed-div
// contract that the server-side strip relies on. editor-ext had ZERO tests
// before this file; this adds the runner + the round-trip coverage.
// The server parses the embed div by matching `data-type="htmlEmbed"` and
// decoding `data-source`; mirror that here so the assertion is exactly what the
// real `htmlToJson` -> htmlEmbed node parse depends on (the node's parseHTML in
// html-embed.ts uses the same selector + decodeHtmlEmbedSource).
const EMBED_DIV_RE = /<div[^>]*\bdata-type="htmlEmbed"[^>]*>/;
function extractEmbedSource(html: string): string | undefined {
const div = EMBED_DIV_RE.exec(html);
if (!div) return undefined;
const enc = /data-source="([^"]*)"/.exec(div[0]);
if (!enc) return undefined;
return decodeHtmlEmbedSource(enc[1]);
}
// Replicates the server's `hasHtmlEmbedNode` decision against the embed *div*
// (the HTML form the server immediately converts to JSON). If this matches, the
// server's JSON-level `hasHtmlEmbedNode` will too, because htmlToJson maps this
// exact div to an htmlEmbed node.
function htmlHasHtmlEmbed(html: string): boolean {
return EMBED_DIV_RE.test(html);
}
describe("markdown <!--html-embed--> import round-trip", () => {
const source = "<script>x</script>";
it("markdownToHtml turns the marker into an htmlEmbed div carrying the source", async () => {
const md = "<!--html-embed:" + encodeHtmlEmbedSource(source) + "-->";
const html = await markdownToHtml(md);
// The marker became the embed div the server recognizes as an htmlEmbed
// node (so the server's hasHtmlEmbedNode would match it after htmlToJson).
expect(htmlHasHtmlEmbed(html)).toBe(true);
// The decoded source is the original script, intact.
expect(extractEmbedSource(html)).toBe(source);
// The raw script is NOT inlined into the HTML — it stays base64 in the
// attribute (the marker itself must not be a direct injection vector).
expect(html).not.toContain("<script>x</script>");
});
it("preserves UTF-8 / special chars in the embedded source", async () => {
const utf8 = '<script>console.log("héllo → 世界")</script>';
const md = "<!--html-embed:" + encodeHtmlEmbedSource(utf8) + "-->";
const html = await markdownToHtml(md);
expect(htmlHasHtmlEmbed(html)).toBe(true);
expect(extractEmbedSource(html)).toBe(utf8);
});
it("an empty marker still produces an htmlEmbed div (empty source)", async () => {
const html = await markdownToHtml("<!--html-embed:-->");
expect(htmlHasHtmlEmbed(html)).toBe(true);
expect(extractEmbedSource(html)).toBe("");
});
it("round-trips htmlToMarkdown -> markdownToHtml preserving the embed marker", async () => {
const encoded = encodeHtmlEmbedSource(source);
// NOTE: turndown drops a *blank* (childless) element before any custom rule
// runs, and the htmlEmbed div is normally childless. The export pipeline
// therefore must give the rule a non-blank div to fire on; we add an inert
// text child here to exercise the real turndown htmlEmbed rule. (A blank
// embed div serializing to "" is asserted separately below as a documented
// edge so this contract drift is visible.)
const startHtml = `<div data-type="htmlEmbed" data-source="${encoded}">x</div>`;
// Export to markdown: the turndown rule emits the <!--html-embed:..-->
// marker (lossless, inert in plain markdown viewers).
const md = htmlToMarkdown(startHtml);
expect(md).toContain("<!--html-embed:" + encoded + "-->");
// Re-import: the marker round-trips back into an embed div with the same
// decoded source — this is the marker <-> embed-div contract the server's
// import strip depends on.
const html = await markdownToHtml(md);
expect(htmlHasHtmlEmbed(html)).toBe(true);
expect(extractEmbedSource(html)).toBe(source);
});
it("documents that a BLANK embed div serializes to empty markdown (turndown drops childless blocks)", () => {
const encoded = encodeHtmlEmbedSource(source);
const blank = `<div data-type="htmlEmbed" data-source="${encoded}"></div>`;
// This pins current behavior so a future change to the turndown rule (e.g.
// making it fire on blank nodes) is caught rather than silently shipping.
expect(htmlToMarkdown(blank)).toBe("");
});
it("the base64 codec itself round-trips (no '<' leaks into the attribute)", () => {
const encoded = encodeHtmlEmbedSource(source);
expect(encoded).not.toContain("<");
expect(decodeHtmlEmbedSource(encoded)).toBe(source);
});
});

View File

@@ -0,0 +1,41 @@
import { Token } from "marked";
interface HtmlEmbedToken {
type: "htmlEmbed";
raw: string;
encoded: string;
}
/**
* Marked extension that rebuilds an `htmlEmbed` node from the HTML comment
* marker produced by the turndown rule (`<!--html-embed:<base64>-->`).
*
* It emits the same marker div the node's `parseHTML` recognizes, so the
* pipeline MD -> HTML -> ProseMirror JSON restores the node (and its
* base64 `data-source`) exactly. We do NOT expand the raw markup here; the
* source stays base64-encoded in the attribute and is only executed by the
* client NodeView.
*/
export const htmlEmbedExtension = {
name: "htmlEmbed",
level: "block" as const,
start(src: string) {
return src.indexOf("<!--html-embed:");
},
tokenizer(src: string): HtmlEmbedToken | undefined {
const rule = /^<!--html-embed:([A-Za-z0-9+/=]*)-->/;
const match = rule.exec(src);
if (match) {
return {
type: "htmlEmbed",
raw: match[0],
encoded: match[1] ?? "",
};
}
},
renderer(token: Token) {
const htmlEmbedToken = token as HtmlEmbedToken;
return `<div data-type="htmlEmbed" data-source="${htmlEmbedToken.encoded}"></div>`;
},
};

View File

@@ -6,6 +6,7 @@ import {
footnoteReferenceExtension,
extractFootnoteDefinitions,
} from "./footnote.marked";
import { htmlEmbedExtension } from "./html-embed.marked";
marked.use({
renderer: {
@@ -43,6 +44,7 @@ marked.use({
mathBlockExtension,
mathInlineExtension,
footnoteReferenceExtension,
htmlEmbedExtension,
],
});

View File

@@ -64,6 +64,7 @@ export function htmlToMarkdown(html: string): string {
mathInline,
mathBlock,
iframeEmbed,
htmlEmbed,
image,
video,
footnoteReference,
@@ -74,6 +75,32 @@ export function htmlToMarkdown(html: string): string {
.replaceAll('<br>', ' ');
}
/**
* Serialize the `htmlEmbed` node to Markdown.
*
* Markdown has no native representation for an arbitrary-HTML block, so we
* preserve the node losslessly as an HTML comment carrying the base64-encoded
* source (the same `data-source` payload the node stores). `markdownToHtml`
* recognizes the same marker and rebuilds the node, so the round-trip
* MD -> HTML -> JSON keeps the source intact. The comment also keeps the raw
* markup inert in the exported `.md` file (it does not render in plain Markdown
* viewers).
*/
function htmlEmbed(turndownService: _TurndownService) {
turndownService.addRule('htmlEmbed', {
filter: function (node: HTMLInputElement) {
return (
node.nodeName === 'DIV' &&
node.getAttribute('data-type') === 'htmlEmbed'
);
},
replacement: function (_content: string, node: HTMLInputElement) {
const encoded = node.getAttribute('data-source') || '';
return `\n\n<!--html-embed:${encoded}-->\n\n`;
},
});
}
function listParagraph(turndownService: _TurndownService) {
turndownService.addRule('paragraph', {
filter: ['p'],

View File

@@ -0,0 +1 @@
export * from "./page-embed";

View File

@@ -0,0 +1,88 @@
import { mergeAttributes, Node } from "@tiptap/core";
import { ReactNodeViewRenderer } from "@tiptap/react";
export interface PageEmbedOptions {
HTMLAttributes: Record<string, any>;
view: any;
}
export interface PageEmbedAttributes {
sourcePageId?: string | null;
}
declare module "@tiptap/core" {
interface Commands<ReturnType> {
pageEmbed: {
insertPageEmbed: (attributes: PageEmbedAttributes) => ReturnType;
};
}
}
/**
* Whole-page live embed. Holds only a `sourcePageId` reference; the node view
* fetches the source page's current content at render time, so the embed stays
* live (no snapshot is stored in the host document). Separate from
* `transclusionReference` (which addresses a single block by `transclusionId`).
*/
export const PageEmbed = Node.create<PageEmbedOptions>({
name: "pageEmbed",
addOptions() {
return {
HTMLAttributes: {},
view: null,
};
},
group: "block",
atom: true,
isolating: true,
selectable: true,
draggable: true,
addAttributes() {
return {
sourcePageId: {
default: null,
parseHTML: (el) => el.getAttribute("data-source-page-id"),
renderHTML: (attrs) =>
attrs.sourcePageId
? { "data-source-page-id": attrs.sourcePageId }
: {},
},
};
},
parseHTML() {
return [{ tag: `div[data-type="${this.name}"]` }];
},
renderHTML({ HTMLAttributes }) {
return [
"div",
mergeAttributes(
{ "data-type": this.name },
this.options.HTMLAttributes,
HTMLAttributes,
),
];
},
addCommands() {
return {
insertPageEmbed:
(attributes) =>
({ commands }) =>
commands.insertContent({
type: this.name,
attrs: attributes,
}),
};
},
addNodeView() {
if (!this.options.view) return null;
this.editor.isInitialized = true;
return ReactNodeViewRenderer(this.options.view);
},
});

View File

@@ -11,6 +11,7 @@
"jsx": "react-jsx",
"sourceMap": true,
"outDir": "./dist",
"rootDir": "./src",
"baseUrl": "./",
"incremental": true,
"skipLibCheck": true,
@@ -20,5 +21,6 @@
"forceConsistentCasingInFileNames": false,
"noFallthroughCasesInSwitch": false
},
"exclude": ["**/*.test.ts", "vitest.config.ts", "dist"]
"include": ["src/**/*"],
"exclude": ["node_modules", "dist", "src/**/*.spec.ts", "src/**/*.test.ts"]
}

View File

@@ -3,6 +3,7 @@ import { defineConfig } from "vitest/config";
export default defineConfig({
test: {
environment: "jsdom",
include: ["src/**/*.test.ts"],
globals: true,
include: ["src/**/*.{test,spec}.ts"],
},
});

View File

@@ -7,12 +7,30 @@ import { createDocmostMcpServer } from "./index.js";
* embedding host (the gitmost NestJS server) bridges its raw Node req/res into
* `handleRequest`. One McpServer + transport is created per MCP session and
* kept alive between requests, keyed by the `mcp-session-id` header.
*
* `config` is EITHER a static `DocmostMcpConfig` (back-compat: stdio + the env
* service account, unchanged) OR a `McpConfigResolver` run once per session at
* `initialize` to bind that session to the request's identity.
*/
export function createMcpHttpHandler(config) {
export function createMcpHttpHandler(config, options = {}) {
// One transport (and one McpServer) per MCP session, keyed by session id.
const transports = {};
// Last activity timestamp per session id, used for idle eviction.
const lastSeen = {};
// Anti-session-fixation: the opaque identity key bound to each session at
// initialize. A later request for that session whose key differs is rejected.
const sessionIdentity = {};
// Write a JSON-RPC error and end the response. Used for the 400/401 paths so
// every early rejection is a well-formed JSON-RPC error, not a torn response.
const sendJsonRpcError = (res, statusCode, code, message) => {
res.statusCode = statusCode;
res.setHeader("Content-Type", "application/json");
res.end(JSON.stringify({
jsonrpc: "2.0",
error: { code, message },
id: null,
}));
};
// Idle session TTL (ms): a session with no activity for this long is evicted.
// Defaults to 30 min; overridable via MCP_SESSION_IDLE_MS.
const idleTtlMs = (() => {
@@ -29,6 +47,7 @@ export function createMcpHttpHandler(config) {
if (now - (lastSeen[sid] ?? 0) > idleTtlMs) {
void transports[sid].close();
delete lastSeen[sid];
delete sessionIdentity[sid];
}
}
}, sweepIntervalMs);
@@ -41,16 +60,23 @@ export function createMcpHttpHandler(config) {
// A new session may only be created by an initialize request without a
// session id.
if (sessionId || !isInitializeRequest(parsedBody)) {
res.statusCode = 400;
res.setHeader("Content-Type", "application/json");
res.end(JSON.stringify({
jsonrpc: "2.0",
error: {
code: -32000,
message: "Bad Request: no valid session ID provided",
},
id: null,
}));
sendJsonRpcError(res, 400, -32000, "Bad Request: no valid session ID provided");
return;
}
// Resolve the per-session config from the request (per-user identity) when
// a resolver was supplied; otherwise use the static config unchanged. The
// resolver may throw (e.g. bad credentials) — surface a clean 401, never
// a created session.
let sessionConfig;
let identity;
try {
sessionConfig =
typeof config === "function" ? await config(req) : config;
if (options.identify)
identity = await options.identify(req);
}
catch (err) {
sendJsonRpcError(res, 401, -32001, err instanceof Error ? err.message : "Unauthorized");
return;
}
transport = new StreamableHTTPServerTransport({
@@ -58,31 +84,46 @@ export function createMcpHttpHandler(config) {
onsessioninitialized: (sid) => {
transports[sid] = transport;
lastSeen[sid] = Date.now();
// Bind the resolved identity to the new session id for anti-fixation.
if (identity !== undefined)
sessionIdentity[sid] = identity;
},
});
transport.onclose = () => {
const sid = transport.sessionId;
if (sid && transports[sid])
delete transports[sid];
if (sid)
delete sessionIdentity[sid];
};
const server = createDocmostMcpServer(config);
const server = createDocmostMcpServer(sessionConfig);
await server.connect(transport);
await transport.handleRequest(req, res, parsedBody);
return;
}
if (!transport) {
res.statusCode = 400;
res.setHeader("Content-Type", "application/json");
res.end(JSON.stringify({
jsonrpc: "2.0",
error: {
code: -32000,
message: "Bad Request: no valid session ID provided",
},
id: null,
}));
sendJsonRpcError(res, 400, -32000, "Bad Request: no valid session ID provided");
return;
}
// Anti-session-fixation: a request reusing an existing session id must
// present credentials/token that resolve to the SAME identity bound at
// initialize, otherwise reject with 401. This prevents hijacking another
// user's established session by replaying its session id with different
// credentials.
if (options.identify && sessionId && sessionId in sessionIdentity) {
let presented;
try {
presented = await options.identify(req);
}
catch (err) {
sendJsonRpcError(res, 401, -32001, err instanceof Error ? err.message : "Unauthorized");
return;
}
if (presented !== sessionIdentity[sessionId]) {
sendJsonRpcError(res, 401, -32001, "Credentials do not match the user that owns this MCP session.");
return;
}
}
// Routing to an existing transport: refresh its idle timestamp.
if (sessionId)
lastSeen[sessionId] = Date.now();

View File

@@ -4,17 +4,71 @@ import { StreamableHTTPServerTransport } from "@modelcontextprotocol/sdk/server/
import { isInitializeRequest } from "@modelcontextprotocol/sdk/types.js";
import { createDocmostMcpServer, DocmostMcpConfig } from "./index.js";
/**
* Per-request config resolver. Run ONCE per MCP session, at the `initialize`
* POST, so the session's DocmostClient is bound to that request's identity
* (e.g. the HTTP-Basic user the embedding host validated). Back-compat: a plain
* `DocmostMcpConfig` object is still accepted (stdio + service account), in
* which case the resolver branch is never taken.
*/
export type McpConfigResolver = (
req: IncomingMessage,
) => DocmostMcpConfig | Promise<DocmostMcpConfig>;
/**
* Optional anti-session-fixation hook. When supplied, it is called on EVERY
* request (init and subsequent) to derive an opaque identity key for the
* presented credentials/token. The key resolved at session init is bound to the
* `mcp-session-id`; a later request whose key differs is rejected with 401, so
* a caller cannot hijack another user's established session by reusing its
* session id with different credentials. The key is opaque to this package (the
* embedding host decides what identity means, e.g. the user's `sub`/email), so
* the package stays generic. Throwing here surfaces as a 401 as well.
*/
export interface McpHttpOptions {
identify?: (req: IncomingMessage) => string | Promise<string>;
}
/**
* Build a stateful Streamable-HTTP handler for the Docmost MCP server. The
* embedding host (the gitmost NestJS server) bridges its raw Node req/res into
* `handleRequest`. One McpServer + transport is created per MCP session and
* kept alive between requests, keyed by the `mcp-session-id` header.
*
* `config` is EITHER a static `DocmostMcpConfig` (back-compat: stdio + the env
* service account, unchanged) OR a `McpConfigResolver` run once per session at
* `initialize` to bind that session to the request's identity.
*/
export function createMcpHttpHandler(config: DocmostMcpConfig) {
export function createMcpHttpHandler(
config: DocmostMcpConfig | McpConfigResolver,
options: McpHttpOptions = {},
) {
// One transport (and one McpServer) per MCP session, keyed by session id.
const transports: Record<string, StreamableHTTPServerTransport> = {};
// Last activity timestamp per session id, used for idle eviction.
const lastSeen: Record<string, number> = {};
// Anti-session-fixation: the opaque identity key bound to each session at
// initialize. A later request for that session whose key differs is rejected.
const sessionIdentity: Record<string, string> = {};
// Write a JSON-RPC error and end the response. Used for the 400/401 paths so
// every early rejection is a well-formed JSON-RPC error, not a torn response.
const sendJsonRpcError = (
res: ServerResponse,
statusCode: number,
code: number,
message: string,
): void => {
res.statusCode = statusCode;
res.setHeader("Content-Type", "application/json");
res.end(
JSON.stringify({
jsonrpc: "2.0",
error: { code, message },
id: null,
}),
);
};
// Idle session TTL (ms): a session with no activity for this long is evicted.
// Defaults to 30 min; overridable via MCP_SESSION_IDLE_MS.
@@ -33,6 +87,7 @@ export function createMcpHttpHandler(config: DocmostMcpConfig) {
if (now - (lastSeen[sid] ?? 0) > idleTtlMs) {
void transports[sid].close();
delete lastSeen[sid];
delete sessionIdentity[sid];
}
}
}, sweepIntervalMs);
@@ -51,17 +106,30 @@ export function createMcpHttpHandler(config: DocmostMcpConfig) {
// A new session may only be created by an initialize request without a
// session id.
if (sessionId || !isInitializeRequest(parsedBody)) {
res.statusCode = 400;
res.setHeader("Content-Type", "application/json");
res.end(
JSON.stringify({
jsonrpc: "2.0",
error: {
code: -32000,
message: "Bad Request: no valid session ID provided",
},
id: null,
}),
sendJsonRpcError(
res,
400,
-32000,
"Bad Request: no valid session ID provided",
);
return;
}
// Resolve the per-session config from the request (per-user identity) when
// a resolver was supplied; otherwise use the static config unchanged. The
// resolver may throw (e.g. bad credentials) — surface a clean 401, never
// a created session.
let sessionConfig: DocmostMcpConfig;
let identity: string | undefined;
try {
sessionConfig =
typeof config === "function" ? await config(req) : config;
if (options.identify) identity = await options.identify(req);
} catch (err) {
sendJsonRpcError(
res,
401,
-32001,
err instanceof Error ? err.message : "Unauthorized",
);
return;
}
@@ -70,33 +138,60 @@ export function createMcpHttpHandler(config: DocmostMcpConfig) {
onsessioninitialized: (sid: string) => {
transports[sid] = transport!;
lastSeen[sid] = Date.now();
// Bind the resolved identity to the new session id for anti-fixation.
if (identity !== undefined) sessionIdentity[sid] = identity;
},
});
transport.onclose = () => {
const sid = transport!.sessionId;
if (sid && transports[sid]) delete transports[sid];
if (sid) delete sessionIdentity[sid];
};
const server = createDocmostMcpServer(config);
const server = createDocmostMcpServer(sessionConfig);
await server.connect(transport);
await transport.handleRequest(req, res, parsedBody);
return;
}
if (!transport) {
res.statusCode = 400;
res.setHeader("Content-Type", "application/json");
res.end(
JSON.stringify({
jsonrpc: "2.0",
error: {
code: -32000,
message: "Bad Request: no valid session ID provided",
},
id: null,
}),
sendJsonRpcError(
res,
400,
-32000,
"Bad Request: no valid session ID provided",
);
return;
}
// Anti-session-fixation: a request reusing an existing session id must
// present credentials/token that resolve to the SAME identity bound at
// initialize, otherwise reject with 401. This prevents hijacking another
// user's established session by replaying its session id with different
// credentials.
if (options.identify && sessionId && sessionId in sessionIdentity) {
let presented: string;
try {
presented = await options.identify(req);
} catch (err) {
sendJsonRpcError(
res,
401,
-32001,
err instanceof Error ? err.message : "Unauthorized",
);
return;
}
if (presented !== sessionIdentity[sessionId]) {
sendJsonRpcError(
res,
401,
-32001,
"Credentials do not match the user that owns this MCP session.",
);
return;
}
}
// Routing to an existing transport: refresh its idle timestamp.
if (sessionId) lastSeen[sessionId] = Date.now();
await transport.handleRequest(req, res, parsedBody);

View File

@@ -0,0 +1,234 @@
// Unit tests for createMcpHttpHandler's config-resolver + anti-fixation hook
// (http.ts). These assert the wrapper contract WITHOUT depending on the MCP
// SDK's full initialize handshake succeeding:
// - a STATIC config is still accepted (back-compat: stdio / service account)
// and never invokes a resolver;
// - a RESOLVER is accepted and is invoked exactly once on a session-init POST;
// - the resolver/identify path runs BEFORE the transport, so a thrown
// resolver error surfaces as a clean 401 and no session is created.
import { test } from "node:test";
import assert from "node:assert/strict";
import { Readable } from "node:stream";
import { createMcpHttpHandler } from "../../build/http.js";
// A minimal initialize JSON-RPC request body (isInitializeRequest checks
// method === "initialize" + jsonrpc + an object params with protocolVersion).
const INIT_BODY = {
jsonrpc: "2.0",
id: 1,
method: "initialize",
params: {
protocolVersion: "2025-03-26",
capabilities: {},
clientInfo: { name: "test", version: "0.0.0" },
},
};
// Fake Node req: a readable stream is fine; we pass parsedBody explicitly so the
// transport never reads the stream, and our resolver short-circuits before that.
function makeReq({ method = "POST", headers = {} } = {}) {
const req = new Readable({ read() {} });
req.method = method;
req.headers = headers;
req.push(null);
return req;
}
// Fake Node res capturing statusCode + body, mimicking just what http.ts uses.
function makeRes() {
const chunks = [];
return {
statusCode: 200,
headers: {},
headersSent: false,
setHeader(k, v) {
this.headers[k.toLowerCase()] = v;
},
end(data) {
if (data) chunks.push(data);
this.headersSent = true;
this.ended = true;
},
body() {
return chunks.join("");
},
};
}
test("static config is accepted and never calls a resolver (back-compat)", async () => {
// A static config object — the stdio / service-account path. A NON-initialize
// POST with no session id must hit the 400 branch deterministically, proving
// the static handler is wired and no resolver is consulted.
const handler = createMcpHttpHandler({
apiUrl: "http://127.0.0.1:3000/api",
email: "svc@example.com",
password: "secret",
});
const req = makeReq({ method: "POST", headers: {} });
const res = makeRes();
await handler.handleRequest(req, res, { jsonrpc: "2.0", method: "ping", id: 9 });
assert.equal(res.statusCode, 400);
assert.match(res.body(), /no valid session ID/);
});
test("resolver is invoked exactly once on a session-init POST", async () => {
let calls = 0;
const handler = createMcpHttpHandler((req) => {
calls += 1;
// Throw a sentinel so we observe invocation without driving the full
// SDK handshake; http.ts turns a resolver throw into a clean 401.
throw new Error("sentinel-from-resolver");
});
const req = makeReq({ method: "POST", headers: {} });
const res = makeRes();
await handler.handleRequest(req, res, INIT_BODY);
assert.equal(calls, 1, "resolver must be called exactly once per init");
assert.equal(res.statusCode, 401);
assert.match(res.body(), /sentinel-from-resolver/);
});
test("resolver is NOT invoked for a non-init POST without a session id", async () => {
let calls = 0;
const handler = createMcpHttpHandler(() => {
calls += 1;
return { apiUrl: "http://127.0.0.1:3000/api", getToken: async () => "t" };
});
const req = makeReq({ method: "POST", headers: {} });
const res = makeRes();
await handler.handleRequest(req, res, { jsonrpc: "2.0", method: "ping", id: 2 });
assert.equal(calls, 0);
assert.equal(res.statusCode, 400);
});
test("identify hook throwing on init surfaces as a clean 401", async () => {
const handler = createMcpHttpHandler(
() => ({ apiUrl: "http://127.0.0.1:3000/api", getToken: async () => "t" }),
{
identify: () => {
throw new Error("bad-identity");
},
},
);
const req = makeReq({ method: "POST", headers: {} });
const res = makeRes();
await handler.handleRequest(req, res, INIT_BODY);
assert.equal(res.statusCode, 401);
assert.match(res.body(), /bad-identity/);
});
// Drive a REAL initialize handshake (over a loopback http server so the SDK's
// StreamableHTTPServerTransport gets genuine Node req/res objects), capture the
// assigned mcp-session-id, then replay subsequent requests to exercise the
// anti-fixation identify comparison: the SAME identity is accepted (routed to
// the transport), a DIFFERENT identity is rejected 401, and crucially the
// per-session config RESOLVER is consulted only ONCE (at init), never on a
// subsequent request — proving subsequent requests do not re-mint the config.
test("subsequent request: SAME identity routes through, DIFFERENT identity is 401, resolver runs once", async () => {
const http = await import("node:http");
let resolverCalls = 0;
let currentIdentity = "user-a";
const handler = createMcpHttpHandler(
() => {
resolverCalls += 1;
return { apiUrl: "http://127.0.0.1:3000/api", getToken: async () => "t" };
},
{ identify: () => currentIdentity },
);
// Loopback server: every request is bridged into the MCP handler with its body
// parsed from JSON, exactly like the embedding host does.
const server = http.createServer((req, res) => {
let raw = "";
req.on("data", (c) => (raw += c));
req.on("end", () => {
const body = raw ? JSON.parse(raw) : undefined;
handler.handleRequest(req, res, body).catch(() => {
if (!res.headersSent) {
res.statusCode = 500;
res.end();
}
});
});
});
await new Promise((r) => server.listen(0, "127.0.0.1", r));
const { port } = server.address();
const call = (headers, body) =>
new Promise((resolve) => {
const r = http.request(
{
host: "127.0.0.1",
port,
method: "POST",
path: "/mcp",
headers: {
"Content-Type": "application/json",
Accept: "application/json, text/event-stream",
...headers,
},
},
(resp) => {
let data = "";
resp.on("data", (c) => (data += c));
resp.on("end", () =>
resolve({
statusCode: resp.statusCode,
sessionId: resp.headers["mcp-session-id"],
body: data,
}),
);
},
);
r.end(JSON.stringify(body));
});
try {
// 1) Establish a session via a real initialize POST (identity = user-a).
const init = await call({}, INIT_BODY);
assert.equal(resolverCalls, 1, "resolver runs exactly once at init");
const sid = init.sessionId;
assert.ok(sid, "initialize must assign an mcp-session-id");
// 2) Subsequent request, SAME identity: not a 401, resolver NOT re-run.
const ok = await call(
{ "mcp-session-id": sid },
{ jsonrpc: "2.0", method: "ping", id: 5 },
);
assert.notEqual(ok.statusCode, 401, "same identity must not be rejected");
assert.equal(resolverCalls, 1, "resolver is NOT re-run on a subsequent request");
// 3) Subsequent request, DIFFERENT identity: rejected 401 (anti-fixation).
currentIdentity = "user-b";
const bad = await call(
{ "mcp-session-id": sid },
{ jsonrpc: "2.0", method: "ping", id: 6 },
);
assert.equal(bad.statusCode, 401, "different identity hijack is rejected");
assert.match(bad.body, /do not match the user/);
assert.equal(resolverCalls, 1, "still no resolver re-run on the rejected request");
} finally {
await new Promise((r) => server.close(r));
}
});
test("unknown existing session id (non-init, with session header) is 400", async () => {
// A request carrying a session id that was never established must not consult
// the resolver or identify hook — it is a plain 400 (no valid session).
let calls = 0;
const handler = createMcpHttpHandler(
() => {
calls += 1;
return { apiUrl: "http://127.0.0.1:3000/api", getToken: async () => "t" };
},
{ identify: () => "x" },
);
const req = makeReq({
method: "POST",
headers: { "mcp-session-id": "does-not-exist" },
});
const res = makeRes();
await handler.handleRequest(req, res, { jsonrpc: "2.0", method: "ping", id: 3 });
assert.equal(res.statusCode, 400);
assert.equal(calls, 0);
});