@contextstream/mcp-server 0.4.81 → 0.4.82

package/README.md

@@ -43,7 +43,7 @@ npx --prefer-online -y @contextstream/mcp-server@latest setup --yes
 ```
 
 <p align="center">
-  <img src="compare1.gif" alt="Side-by-side comparison: an AI coding assistant with ContextStream memory and semantic search vs. without" width="700" />
+  <img src="contextstream-mcp-comparison.gif" alt="Animated comparison of an AI coding assistant working without ContextStream versus with ContextStream project memory, search, and graph context" width="600" />
 </p>
 
 ---
@@ -72,6 +72,21 @@ It works with any MCP client: Claude Code, Cursor, VS Code + GitHub Copilot, Win
 
 ---
 
+## Which ContextStream runtime am I using?
+
+ContextStream has independent release lines. A `0.5.x` hosted version and a `0.4.x` npm version are different runtimes, not evidence that either updater is broken.
+
+| Runtime | How to identify it | Version line | Canonical release metadata and notes |
+|---|---|---|---|
+| **Hosted MCP** | Your editor config uses `https://mcp.contextstream.io/mcp`. `help(action="version")` reports `runtime_type: rust-mcp`. | Rust MCP `0.5.x` | `help(action="version")` returns release notes inline when published and links the [machine-readable R2 manifest](https://pub-68429b9f7857416c9484b75bf1887b96.r2.dev/mcp/latest/version.json). |
+| **Installed Rust MCP** | Your editor launches a `contextstream-mcp` binary installed by `https://contextstream.io/scripts/mcp.sh`; run `contextstream-mcp --version`. | Rust MCP `0.5.x` | The same [machine-readable R2 manifest](https://pub-68429b9f7857416c9484b75bf1887b96.r2.dev/mcp/latest/version.json); `help(action="version")` maps this runtime to that manifest. |
+| **Legacy npm MCP (this repository)** | Your editor launches `npx ... @contextstream/mcp-server` or a global npm install. `help(action="version")` reports `runtime_type: legacy-typescript-mcp`; `npm list -g @contextstream/mcp-server` shows the installed package. | TypeScript MCP `0.4.x` | [GitHub releases](https://github.com/contextstream/mcp-server/releases) and this repository's [CHANGELOG.md](CHANGELOG.md). |
+| **ContextStream Desktop** | Check the app's About/update UI. Desktop may run the local sync bridge, but that does not change the MCP runtime configured in your editor. | Desktop `0.3.x` (independent) | The in-app updater and the public [Desktop version JSON](https://api.contextstream.io/api/v1/desktop/version), which includes `release_notes` and platform downloads. |
+
+The `version` reported by `help(action="version")` is always the MCP process serving that tool call. Desktop's version is separate, even when Desktop added or indexed the local repository.
+
+---
+
 ## Why do AI coding assistants forget everything?
 
 Because every conversation starts from zero. Your AI re-reads the same files, re-derives the same architecture, repeats last week's mistake, and loses the thread the moment the context window compacts. ContextStream fixes the whole class of problem:
@@ -130,23 +145,33 @@ Anything that speaks the Model Context Protocol can connect — the table just s
 
 ---
 
-## The tools your AI gets
+## Tools
 
-36 tools in the default surface, organized as consolidated domains so they cost ~75% fewer tokens than individual registrations:
+36 tools in the default surface, organized as consolidated domains so they cost ~75% fewer tokens than individual registrations. The tools your AI gets:
 
-```
-init / context   → workspace state + the right context on every message
-search           → semantic, hybrid, keyword, pattern, exhaustive, refactor modes
-memory           → events, decisions, docs, runbooks, tasks, todos, diagrams, transcripts
-session          → capture decisions & lessons, recall past sessions, plans, retroactive capture
-qa               → grounded Q&A over your workspace knowledge base, with citations
-graph            → dependencies, impact analysis, circular deps, unused code
-capsule          → portable, shareable context snapshots for agent handoffs
-entity           → tickets, incidents, releases, sprints, OKRs, risks
-project / workspace / skill / media / vcs / reminder / integration / help
-```
+- **init** / **context** — workspace state + the right context on every message
+- **search** — semantic, hybrid, keyword, pattern, exhaustive, refactor modes
+- **memory** — events, decisions, docs, runbooks, tasks, todos, diagrams, transcripts
+- **session** — capture decisions & lessons, recall past sessions, plans, retroactive capture
+- **qa** — grounded Q&A over your workspace knowledge base, with citations
+- **graph** — dependencies, impact analysis, circular deps, unused code
+- **capsule** — portable, shareable context snapshots for agent handoffs
+- **entity** — tickets, incidents, releases, sprints, OKRs, risks
+- **project** / **workspace** — project indexing, scope, and workspace management
+- **skill** — reusable instruction + action bundles, portable across tools
+- **media** — index and search images, video, audio, and documents
+- **vcs** / **reminder** / **integration** / **help** — repo links, reminders, integrations, diagnostics
+
+Plus focused write tools (`capture_plan`, `memory_create_doc`, `session_capture_lesson`, …) so agents that display tool names show *what* they're doing. Your AI uses all of this automatically — you just code.
+
+### Daily Recaps
+
+Daily Recaps are generated around **23:00 in your configured timezone** when there is enough activity. They are not triggered by closing an editor, changing an MCP `session_id`, or starting a new chat, so long-lived VS Code/Copilot connections do not suppress the nightly job.
+
+- `session(action="list_recaps", workspace_id="<uuid>", limit=30)` lists completed recaps newest-first with `recap_date` and `generated_at` timestamps.
+- `session(action="trigger_recap", workspace_id="<uuid>")` queues an asynchronous manual recap. Call `list_recaps` afterward to verify completion.
 
-Plus focused write tools (`capture_plan`, `memory_create_doc`, …) so agents that display tool names show *what* they're doing. Your AI uses all of this automatically — you just code.
+These actions use the same recap history and generation service as the dashboard.
 
 ---
 

package/dist/hooks/auto-rules.js

@@ -5,11 +5,20 @@ var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
 var __getOwnPropNames = Object.getOwnPropertyNames;
 var __getProtoOf = Object.getPrototypeOf;
 var __hasOwnProp = Object.prototype.hasOwnProperty;
-var __esm = (fn, res) => function __init() {
-  return fn && (res = (0, fn[__getOwnPropNames(fn)[0]])(fn = 0)), res;
+var __esm = (fn, res, err) => function __init() {
+  if (err) throw err[0];
+  try {
+    return fn && (res = (0, fn[__getOwnPropNames(fn)[0]])(fn = 0)), res;
+  } catch (e) {
+    throw err = [e], e;
+  }
 };
 var __commonJS = (cb, mod) => function __require() {
-  return mod || (0, cb[__getOwnPropNames(cb)[0]])((mod = { exports: {} }).exports, mod), mod.exports;
+  try {
+    return mod || (0, cb[__getOwnPropNames(cb)[0]])((mod = { exports: {} }).exports, mod), mod.exports;
+  } catch (e) {
+    throw mod = 0, e;
+  }
 };
 var __export = (target, all) => {
   for (var name in all)
@@ -492,39 +501,96 @@ var require_ignore = __commonJS({
 // src/ignore.ts
 import * as fs from "fs";
 import * as path from "path";
+function isInvariantIgnoredPath(pathname) {
+  const normalized = pathname.replace(/\\/g, "/").replace(/^\.\//, "");
+  const segments = normalized.split("/").filter(Boolean);
+  if (segments.some((segment) => INVARIANT_IGNORED_DIRECTORIES.has(segment.toLowerCase()))) {
+    return true;
+  }
+  const basename = (segments.at(-1) || "").toLowerCase();
+  if (INVARIANT_SECRET_FILENAMES.has(basename)) return true;
+  if (INVARIANT_SECRET_SUFFIXES.some((suffix) => basename.endsWith(suffix))) return true;
+  return (basename === ".env" || basename.startsWith(".env.")) && !ENV_TEMPLATE_ALLOWLIST.has(basename);
+}
 async function loadIgnorePatterns(projectRoot) {
   const ig = (0, import_ignore.default)();
   const patterns = [...DEFAULT_IGNORE_PATTERNS];
   ig.add(DEFAULT_IGNORE_PATTERNS);
-  const ignoreFilePath = path.join(projectRoot, IGNORE_FILENAME);
   let hasUserPatterns = false;
-  try {
-    const content = await fs.promises.readFile(ignoreFilePath, "utf-8");
-    const userPatterns = content.split("\n").map((line) => line.trim()).filter((line) => line && !line.startsWith("#"));
-    if (userPatterns.length > 0) {
-      ig.add(userPatterns);
-      patterns.push(...userPatterns);
-      hasUserPatterns = true;
+  for (const filename of IGNORE_FILENAMES) {
+    try {
+      const content = await fs.promises.readFile(path.join(projectRoot, filename), "utf-8");
+      const userPatterns = content.split("\n").map((line) => line.trim()).filter((line) => line && !line.startsWith("#"));
+      if (userPatterns.length > 0) {
+        ig.add(userPatterns);
+        patterns.push(...userPatterns);
+        hasUserPatterns = true;
+      }
+    } catch {
     }
-  } catch {
   }
   return {
-    ignores: (pathname) => ig.ignores(pathname),
+    ignores: (pathname) => isInvariantIgnoredPath(pathname) || ig.ignores(pathname),
     patterns,
     hasUserPatterns
   };
 }
-var import_ignore, IGNORE_FILENAME, DEFAULT_IGNORE_PATTERNS;
+var import_ignore, IGNORE_FILENAMES, INVARIANT_IGNORED_DIRECTORIES, INVARIANT_SECRET_FILENAMES, INVARIANT_SECRET_SUFFIXES, ENV_TEMPLATE_ALLOWLIST, DEFAULT_IGNORE_PATTERNS;
 var init_ignore = __esm({
   "src/ignore.ts"() {
     "use strict";
     import_ignore = __toESM(require_ignore(), 1);
-    IGNORE_FILENAME = ".contextstream/ignore";
+    IGNORE_FILENAMES = [".gitignore", ".contextignore", ".contextstream/ignore"];
+    INVARIANT_IGNORED_DIRECTORIES = /* @__PURE__ */ new Set([
+      ".contextstream",
+      ".claude",
+      ".codex",
+      ".cursor",
+      ".windsurf",
+      ".roo",
+      ".kilocode",
+      ".ssh",
+      ".aws",
+      ".gnupg",
+      ".kube",
+      ".docker"
+    ]);
+    INVARIANT_SECRET_FILENAMES = /* @__PURE__ */ new Set([
+      ".npmrc",
+      ".pypirc",
+      ".netrc",
+      ".pgpass",
+      ".git-credentials",
+      "id_rsa",
+      "id_ed25519",
+      "id_ecdsa",
+      "known_hosts",
+      "credentials.json",
+      "serviceaccountkey.json",
+      "auth.json"
+    ]);
+    INVARIANT_SECRET_SUFFIXES = [".pem", ".key", ".p12", ".pfx", ".jks"];
+    ENV_TEMPLATE_ALLOWLIST = /* @__PURE__ */ new Set([
+      ".env.example",
+      ".env.sample",
+      ".env.template",
+      ".env.dist",
+      ".env.defaults"
+    ]);
     DEFAULT_IGNORE_PATTERNS = [
       // Version control
       ".git/",
       ".svn/",
       ".hg/",
+      // ContextStream and AI editor/agent state. These directories can contain
+      // credentials, generated rules, caches, and nested worktrees.
+      ".contextstream/",
+      ".claude/",
+      ".codex/",
+      ".cursor/",
+      ".windsurf/",
+      ".roo/",
+      ".kilocode/",
       // Package managers / dependencies
       "node_modules/",
       "vendor/",
@@ -555,6 +621,12 @@ var init_ignore = __esm({
       ".idea/",
       ".vscode/",
       ".vs/",
+      // Credential stores
+      ".ssh/",
+      ".aws/",
+      ".gnupg/",
+      ".kube/",
+      ".docker/",
       // Coverage
       "coverage/",
       ".coverage/",
@@ -568,6 +640,22 @@ var init_ignore = __esm({
       "composer.lock",
       "*.min.js",
       "*.min.css",
+      "*.pem",
+      "*.key",
+      "*.p12",
+      "*.pfx",
+      "*.jks",
+      ".npmrc",
+      ".pypirc",
+      ".netrc",
+      ".pgpass",
+      ".git-credentials",
+      "id_rsa",
+      "id_ed25519",
+      "id_ecdsa",
+      "credentials.json",
+      "serviceAccountKey.json",
+      "auth.json",
       // OS files
       ".DS_Store",
       "Thumbs.db"
@@ -595,6 +683,15 @@ import { exec } from "child_process";
 import { promisify } from "util";
 import * as os from "os";
 import * as crypto from "crypto";
+function shouldSkipSensitiveFile(name) {
+  const lower = name.toLowerCase();
+  if (SECRET_FILENAMES.has(lower)) return true;
+  if (SECRET_FILE_SUFFIXES.some((suffix) => lower.endsWith(suffix))) return true;
+  if ((lower === ".env" || lower.startsWith(".env.")) && !ENV_TEMPLATE_ALLOWLIST2.has(lower)) {
+    return true;
+  }
+  return false;
+}
 function getMachineId() {
   const hostname2 = os.hostname();
   const hash = crypto.createHash("sha256").update(hostname2).digest("hex");
@@ -701,6 +798,7 @@ function clearGitContextCache() {
   gitContextCache.clear();
 }
 async function readFilesFromDirectory(rootPath, options = {}) {
+  if (isInvariantIgnoredPath(path2.resolve(rootPath))) return [];
   const maxFiles = options.maxFiles ?? MAX_FILES_PER_BATCH;
   const maxFileSize = options.maxFileSize ?? MAX_FILE_SIZE;
   const files = [];
@@ -725,6 +823,7 @@ async function readFilesFromDirectory(rootPath, options = {}) {
         await walkDir(fullPath, relPath);
       } else if (entry.isFile()) {
         if (IGNORE_FILES.has(entry.name)) continue;
+        if (shouldSkipSensitiveFile(entry.name)) continue;
         if (ig.ignores(relPath)) continue;
         const ext = entry.name.split(".").pop()?.toLowerCase() ?? "";
         if (!CODE_EXTENSIONS.has(ext)) continue;
@@ -760,6 +859,7 @@ async function readFilesFromDirectory(rootPath, options = {}) {
   return files;
 }
 async function* readAllFilesInBatches(rootPath, options = {}) {
+  if (isInvariantIgnoredPath(path2.resolve(rootPath))) return;
   const maxBatchBytes = options.maxBatchBytes ?? MAX_BATCH_BYTES;
   const largeFileThreshold = options.largeFileThreshold ?? LARGE_FILE_THRESHOLD;
   const maxFilesPerBatch = options.maxFilesPerBatch ?? options.batchSize ?? MAX_FILES_PER_BATCH;
@@ -785,6 +885,7 @@ async function* readAllFilesInBatches(rootPath, options = {}) {
         yield* walkDir(fullPath, relPath);
       } else if (entry.isFile()) {
         if (IGNORE_FILES.has(entry.name)) continue;
+        if (shouldSkipSensitiveFile(entry.name)) continue;
         if (ig.ignores(relPath)) continue;
         const ext = entry.name.split(".").pop()?.toLowerCase() ?? "";
         if (!CODE_EXTENSIONS.has(ext)) continue;
@@ -845,6 +946,7 @@ async function* readAllFilesInBatches(rootPath, options = {}) {
   }
 }
 async function* readChangedFilesInBatches(rootPath, sinceTimestamp, options = {}) {
+  if (isInvariantIgnoredPath(path2.resolve(rootPath))) return;
   const maxBatchBytes = options.maxBatchBytes ?? MAX_BATCH_BYTES;
   const largeFileThreshold = options.largeFileThreshold ?? LARGE_FILE_THRESHOLD;
   const maxFilesPerBatch = options.maxFilesPerBatch ?? options.batchSize ?? MAX_FILES_PER_BATCH;
@@ -873,6 +975,7 @@ async function* readChangedFilesInBatches(rootPath, sinceTimestamp, options = {}
         yield* walkDir(fullPath, relPath);
       } else if (entry.isFile()) {
         if (IGNORE_FILES.has(entry.name)) continue;
+        if (shouldSkipSensitiveFile(entry.name)) continue;
         if (ig.ignores(relPath)) continue;
         const ext = entry.name.split(".").pop()?.toLowerCase() ?? "";
         if (!CODE_EXTENSIONS.has(ext)) continue;
@@ -939,6 +1042,7 @@ async function* readChangedFilesInBatches(rootPath, sinceTimestamp, options = {}
   );
 }
 async function countIndexableFiles(rootPath, options = {}) {
+  if (isInvariantIgnoredPath(path2.resolve(rootPath))) return { count: 0, stopped: false };
   const maxFiles = options.maxFiles ?? 1;
   const maxFileSize = options.maxFileSize ?? MAX_FILE_SIZE;
   const ig = options.ignoreInstance ?? await loadIgnorePatterns(rootPath);
@@ -969,6 +1073,7 @@ async function countIndexableFiles(rootPath, options = {}) {
         await walkDir(fullPath, relPath);
       } else if (entry.isFile()) {
         if (IGNORE_FILES.has(entry.name)) continue;
+        if (shouldSkipSensitiveFile(entry.name)) continue;
         if (ig.ignores(relPath)) continue;
         const ext = entry.name.split(".").pop()?.toLowerCase() ?? "";
         if (!CODE_EXTENSIONS.has(ext)) continue;
@@ -1053,7 +1158,7 @@ function deleteHashManifest(projectId) {
   } catch {
   }
 }
-var execAsync, CODE_EXTENSIONS, IGNORE_DIRS, IGNORE_FILES, MAX_FILE_SIZE, MAX_BATCH_BYTES, LARGE_FILE_THRESHOLD, MAX_FILES_PER_BATCH, gitContextCache;
+var execAsync, CODE_EXTENSIONS, IGNORE_DIRS, IGNORE_FILES, SECRET_FILENAMES, SECRET_FILE_SUFFIXES, ENV_TEMPLATE_ALLOWLIST2, MAX_FILE_SIZE, MAX_BATCH_BYTES, LARGE_FILE_THRESHOLD, MAX_FILES_PER_BATCH, gitContextCache;
 var init_files = __esm({
   "src/files.ts"() {
     "use strict";
@@ -1129,6 +1234,13 @@ var init_files = __esm({
     IGNORE_DIRS = /* @__PURE__ */ new Set([
       "node_modules",
       ".git",
+      ".contextstream",
+      ".claude",
+      ".codex",
+      ".cursor",
+      ".windsurf",
+      ".roo",
+      ".kilocode",
       ".svn",
       ".hg",
       "target",
@@ -1156,7 +1268,18 @@ var init_files = __esm({
       "obj",
       ".idea",
       ".vscode",
-      ".vs"
+      ".vs",
+      ".npm",
+      ".yarn",
+      ".nyc_output",
+      // Credential directories are invariant exclusions. Their contents often
+      // have ordinary-looking names (config, credentials, known_hosts), so a
+      // filename-only filter is not sufficient.
+      ".ssh",
+      ".aws",
+      ".gnupg",
+      ".kube",
+      ".docker"
     ]);
     IGNORE_FILES = /* @__PURE__ */ new Set([
       ".DS_Store",
@@ -1171,6 +1294,28 @@ var init_files = __esm({
       "Gemfile.lock",
       "composer.lock"
     ]);
+    SECRET_FILENAMES = /* @__PURE__ */ new Set([
+      ".npmrc",
+      ".pypirc",
+      ".netrc",
+      ".pgpass",
+      ".git-credentials",
+      "id_rsa",
+      "id_ed25519",
+      "id_ecdsa",
+      "known_hosts",
+      "credentials.json",
+      "serviceaccountkey.json",
+      "auth.json"
+    ]);
+    SECRET_FILE_SUFFIXES = [".pem", ".key", ".p12", ".pfx", ".jks"];
+    ENV_TEMPLATE_ALLOWLIST2 = /* @__PURE__ */ new Set([
+      ".env.example",
+      ".env.sample",
+      ".env.template",
+      ".env.dist",
+      ".env.defaults"
+    ]);
     MAX_FILE_SIZE = 5 * 1024 * 1024;
     MAX_BATCH_BYTES = 1024 * 1024;
     LARGE_FILE_THRESHOLD = 2 * 1024 * 1024;