@cachly-dev/mcp-server 0.5.35 → 0.5.37

package/README.md

@@ -1,6 +1,6 @@
 # 🧠 cachly AI Brain — MCP Server
 
-> **Persistent memory for Claude Code, Cursor, GitHub Copilot & Windsurf.**  
+> **Persistent memory for Claude Code, Cursor, GitHub Copilot, Windsurf, Cline & Zed.**  
 > Your AI remembers every lesson, every fix, every architecture decision — forever.
 
 <p align="center">
@@ -45,9 +45,15 @@ npx @cachly-dev/mcp-server@latest setup
 The interactive wizard:
 1. Signs you in (free, no credit card required)
 2. Picks or creates your AI Brain instance
-3. **Auto-detects** Cursor, Windsurf, VS Code, Claude Code, Continue.dev
+3. **Auto-detects** Cursor, Windsurf, VS Code, Claude Code, Continue.dev, **Cline & Zed**
 4. Writes the correct MCP config for every detected editor
 5. Creates `CLAUDE.md` with memory rules pre-filled
+6. Shows your Brain Health snapshot right in the terminal
+
+**Index your project** into the Brain for instant semantic search:
+```bash
+npx @cachly-dev/mcp-server@latest index .
+```
 
 **Result:** Your AI remembers everything. Always.
 

package/dist/index.d.ts

@@ -6,10 +6,33 @@
  * Keeps the index small and scores meaningful.
  */
 declare const STOPWORDS: Set<string>;
+/**
+ * Convert a katakana string to Hepburn romaji.
+ * Handles digraphs, geminate consonants (ッ), and long vowel marks (ー).
+ */
+declare function katakanaToRomaji(kata: string): string;
+/**
+ * Arabic light stemmer — strips definite article and common prefix particles.
+ * Handles: ال (al-), و (wa-), ب (bi-), ل (li-), ف (fa-), ك (ka-).
+ * Only strips when the result is still ≥ 3 chars to avoid over-stemming.
+ */
+declare function arabicLightStem(word: string): string;
+declare const CROSS_LINGUAL_MAP: Map<string, string[]>;
+/**
+ * Expand a token to its cross-lingual synonyms.
+ * Returns synonyms to be added to the token stream (pre-tokenized).
+ * CJK synonyms are NOT pre-bigram'd here — caller handles that.
+ */
+declare function expandCrossLingual(token: string): string[];
 /**
  * Tokenize text into meaningful keywords.
- * Handles accented Latin characters and CJK (Chinese, Japanese, Korean).
- * CJK text is split into character bigrams (the standard for indexing logographic languages).
+ * Handles accented Latin, RTL (Arabic, Hebrew), and CJK (Chinese, Japanese, Korean).
+ *
+ * Features:
+ *  • CJK → character bigrams
+ *  • Katakana → additionally emits Hepburn romaji tokens (enables romaji queries)
+ *  • Arabic/Hebrew → word tokenization with light Arabic stemming
+ *  • All scripts → cross-lingual synonym expansion (EN↔JA↔ZH↔KO↔AR↔HE)
  */
 declare function tokenize(text: string): string[];
 /**
@@ -33,4 +56,4 @@ declare function levenshtein(a: string, b: string): number;
 declare function extractTimestamp(content: string): number | undefined;
 /** Recency multiplier: 1.0 for now, 0.5 after half-life days, exponential decay. */
 declare function recencyBoost(timestampMs: number | undefined): number;
-export { tokenize, splitMultiQuery, levenshtein, recencyBoost, extractTimestamp, STOPWORDS };
+export { tokenize, splitMultiQuery, levenshtein, recencyBoost, extractTimestamp, STOPWORDS, katakanaToRomaji, arabicLightStem, expandCrossLingual, CROSS_LINGUAL_MAP };