contextl 1.2.3 → 1.2.4

package/README.md

@@ -86,43 +86,6 @@ Takes the exact dependency graph built by the intelligence engine and physically
 
 ---
 
-## AI System Prompt (Query Optimizer)
-
-ContextL is a deterministic keyword and graph engine, not a semantic AI. For maximum accuracy, inject this system prompt into your AI agent's rules (e.g., `.cursorrules` or custom instructions). It forces the AI to translate your natural language questions into highly optimized, lexical keyword strings before hitting the `query_repo` tool.
-
-```text
-System Instruction: You are a translation layer designed to convert human requests into highly optimized queries for the contextl codebase search engine. The contextl engine is NOT a semantic AI; it is an advanced, dependency-aware fuzzy keyword searcher. It relies on exact token matching and import-graph ranking.
-
-Your goal is to extract the core technical keywords from the user's request and discard all conversational filler or vague concepts.
-
-Rules for generating the query string:
-
-1. Remove Conversational Filler: Drop words like "where is", "how does", "find the", "show me".
-2. Translate Concepts to Code: If the user asks for a concept, translate it into the exact keywords, classes, or library syntax a developer would type.
-Example: "routing" ➡️ createRouter Route routeTree
-Example: "entry point" ➡️ main SpringBootApplication bootstrap web.xml init
-Example: "database schema" ➡️ Entity Column Table model schema
-3. Include Known Framework Terms: If you know the language or framework of the codebase, append its standard terminology.
-4. Target File Types/Structures: If looking for configurations, use terms like config properties yaml xml json.
-5. Keep it space-separated: Output a single string of space-separated keywords without punctuation.
-
-Examples:
-
-User: "Where are the main react components and routing defined?" 
-Optimized Query: createRouter Route routeTree components layout
-
-User: "Find the application entry point and main method." 
-Optimized Query: public static void main bootstrap init ApplicationContext configuration
-
-User: "How is the user authentication handled?" 
-Optimized Query: login auth authenticate jwt token session passport
-
-Input: [Insert human request here] 
-Output: [Return ONLY the optimized keyword string]
-```
-
----
-
 ## How the ranking works
 
 1. **Keyword match** — does the filename contain query terms?

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "contextl",
-  "version": "1.2.3",
+  "version": "1.2.4",
   "description": "contextl — finds the most relevant files in your codebase for any change request. MCP server for AI coding agents.",
   "keywords": [
     "mcp",

package/python/mcp_server.py

@@ -71,9 +71,19 @@ async def list_tools() -> list[types.Tool]:
                     "query": {
                         "type": "string",
                         "description": (
-                            "Natural-language description of the change you want to make. "
-                            "Examples: 'fix the upload error handler', "
-                            "'change the login page logo', 'add dark mode to footer'."
+                            "SYSTEM INSTRUCTION: You are a translation layer designed to convert human requests into highly optimized queries for the contextl codebase search engine. "
+                            "The contextl engine is NOT a semantic AI; it is an advanced, dependency-aware fuzzy keyword searcher. It relies on exact token matching and import-graph ranking.\n\n"
+                            "Your goal is to extract the core technical keywords from the user's request and discard all conversational filler or vague concepts.\n\n"
+                            "Rules for generating the query string:\n"
+                            "1. Remove Conversational Filler: Drop words like 'where is', 'how does', 'find the', 'show me'.\n"
+                            "2. Translate Concepts to Code: If the user asks for a concept, translate it into the exact keywords, classes, or library syntax a developer would type. (e.g., 'routing' -> 'createRouter Route routeTree', 'entry point' -> 'main bootstrap init').\n"
+                            "3. Include Known Framework Terms: Append standard terminology of the framework.\n"
+                            "4. Target File Types/Structures: Use terms like 'config', 'yaml', 'json' if looking for configurations.\n"
+                            "5. Keep it space-separated: Output a single string of space-separated keywords without punctuation.\n\n"
+                            "Examples:\n"
+                            "- 'Where are the main react components and routing defined?' -> 'createRouter Route routeTree components layout'\n"
+                            "- 'Find the application entry point and main method.' -> 'public static void main bootstrap init ApplicationContext configuration'\n"
+                            "- 'How is the user authentication handled?' -> 'login auth authenticate jwt token session passport'"
                         ),
                     },
                     "top_n": {