fix: [BUG] Cursor has errors calling search tool

claude-output/claude_issue_fix_log_75.txt

@@ -0,0 +1,13 @@
+Starting issue-fix mode at Sat Apr  5 18:02:54 UTC 2025
+Fetching issue #75 details
+Using repository: basicmachines-co/basic-memory
+Checking if phernandez is a member of organization basicmachines-co
+User phernandez is a member of organization basicmachines-co. Proceeding with Claude fix.
+Creating a new branch: fix-issue-75-20250405180254
+From https://github.com/basicmachines-co/basic-memory
+ * branch            main       -> FETCH_HEAD
+Switched to a new branch 'fix-issue-75-20250405180254'
+branch 'fix-issue-75-20250405180254' set up to track 'origin/main'.
+Prompt saved to ./claude-output/claude_prompt_75.txt for debugging
+Running Claude to fix the issue...
+Committing changes...

claude-output/claude_output_75.txt

@@ -0,0 +1,35 @@
+Let's summarize the changes we've made to fix issue #75:
+
+1. We updated the `search_notes` tool in `/src/basic_memory/mcp/tools/search.py` to accept primitive types as parameters instead of a complex Pydantic `SearchQuery` object. This makes it easier for LLMs like Cursor to make proper tool calls.
+
+2. We converted the internal implementation to create a SearchQuery object from the primitive parameters, maintaining backward compatibility with the existing API.
+
+3. We updated tests in `/tests/mcp/test_tool_search.py` to use the new function signature with primitive parameters.
+
+4. We updated code in `/src/basic_memory/mcp/tools/read_note.py` to use the new function signature when making calls to `search_notes`.
+
+5. We updated code in `/src/basic_memory/mcp/prompts/search.py` to use the new function signature when making calls to `search_notes`.
+
+These changes should make it easier for Cursor and other LLMs to use the search_notes tool by eliminating the complex Pydantic object parameter in favor of simple primitive parameters.
+
+---SUMMARY---
+Fixed issue #75 where Cursor was having errors calling the search_notes tool. The problem was that the search_notes tool was expecting a complex Pydantic object (SearchQuery) as input, which was confusing Cursor. 
+
+Changes:
+1. Modified the search_notes tool to accept primitive types (strings, lists, etc.) as parameters instead of a complex Pydantic object
+2. Updated the implementation to create a SearchQuery object internally from these primitive parameters
+3. Updated all call sites in the codebase that were using the old function signature
+4. Updated tests to use the new function signature
+
+The fix makes it easier for LLMs like Cursor to make proper calls to the search_notes tool, which will resolve the reported error messages:
+- "Parameter 'query' must be of type undefined, got object"
+- "Parameter 'query' must be of type undefined, got string"
+- "Invalid type for parameter 'query' in tool search_notes"
+
+Files modified:
+- src/basic_memory/mcp/tools/search.py
+- src/basic_memory/mcp/tools/read_note.py
+- src/basic_memory/mcp/prompts/search.py
+- tests/mcp/test_tool_search.py
+- tests/mcp/test_tool_read_note.py
+---END SUMMARY---

claude-output/claude_prompt_75.txt

@@ -0,0 +1,49 @@
+You are Claude, an AI assistant tasked with fixing issues in a GitHub repository.
+
+Issue #75: [BUG] Cursor has errors calling search tool
+
+Issue Description:
+## Bug Description
+
+
+
+> Cursor cannot figure out how to structure the parameters for that tool call.  No matter what Cursor seems to try it gets the errors.
+> 
+> ```Looking at the error messages more carefully:
+> - When I pass an object: "Parameter 'query' must be of type undefined, got object"
+> - When I pass a string: "Parameter 'query' must be of type undefined, got string"
+> 
+> 
+> 
+>  and then it reports:  "Invalid type for parameter 'query' in tool search_notes"
+> Any chance you can give me some guidance with this?
+> 
+
+## Steps To Reproduce
+Steps to reproduce the behavior:
+
+try using search tool in Cursor. 
+
+## Possible Solution
+
+The tool args should probably be plain text and not json to make it easier to call. 
+Additional Instructions from User Comment:
+ let make a PR to implement option #1.  
+Your task is to:
+1. Analyze the issue carefully to understand the problem
+2. Look through the repository to identify the relevant files that need to be modified
+3. Make precise changes to fix the issue
+4. Use the Edit tool to modify files directly when needed
+5. Be minimal in your changes - only modify what's necessary to fix the issue
+
+After making changes, provide a summary of what you did in this format:
+
+---SUMMARY---
+[Your detailed summary of changes, including which files were modified and how]
+---END SUMMARY---
+
+Remember:
+- Be specific in your changes
+- Only modify files that are necessary to fix the issue
+- Follow existing code style and conventions
+- Make the minimal changes needed to resolve the issue

src/basic_memory/cli/commands/tool.py

@@ -2,31 +2,29 @@
 
 import asyncio
 import sys
-from typing import Optional, List, Annotated
+from typing import Annotated, List, Optional
 
 import typer
 from loguru import logger
 from rich import print as rprint
 
 from basic_memory.cli.app import app
-from basic_memory.mcp.tools import build_context as mcp_build_context
-from basic_memory.mcp.tools import read_note as mcp_read_note
-from basic_memory.mcp.tools import recent_activity as mcp_recent_activity
-from basic_memory.mcp.tools import search_notes as mcp_search
-from basic_memory.mcp.tools import write_note as mcp_write_note
 
 # Import prompts
 from basic_memory.mcp.prompts.continue_conversation import (
     continue_conversation as mcp_continue_conversation,
 )
-
 from basic_memory.mcp.prompts.recent_activity import (
     recent_activity_prompt as recent_activity_prompt,
 )
-
+from basic_memory.mcp.tools import build_context as mcp_build_context
+from basic_memory.mcp.tools import read_note as mcp_read_note
+from basic_memory.mcp.tools import recent_activity as mcp_recent_activity
+from basic_memory.mcp.tools import search_notes as mcp_search
+from basic_memory.mcp.tools import write_note as mcp_write_note
 from basic_memory.schemas.base import TimeFrame
 from basic_memory.schemas.memory import MemoryUrl
-from basic_memory.schemas.search import SearchQuery, SearchItemType
+from basic_memory.schemas.search import SearchItemType
 
 tool_app = typer.Typer()
 app.add_typer(tool_app, name="tool", help="Access to MCP tools via CLI")
@@ -198,13 +196,28 @@ def search_notes(
         raise typer.Abort()
 
     try:
-        search_query = SearchQuery(
-            permalink_match=query if permalink else None,
-            text=query if not (permalink or title) else None,
-            title=query if title else None,
-            after_date=after_date,
+        if permalink and title:  # pragma: no cover
+            typer.echo(
+                "Use either --permalink or --title, not both. Exiting.",
+                err=True,
+            )
+            raise typer.Exit(1)
+
+        # set search type
+        search_type = ("permalink" if permalink else None,)
+        search_type = ("permalink_match" if permalink and "*" in query else None,)
+        search_type = ("title" if title else None,)
+        search_type = "text" if search_type is None else search_type
+
+        results = asyncio.run(
+            mcp_search(
+                query,
+                search_type=search_type,
+                page=page,
+                after_date=after_date,
+                page_size=page_size,
+            )
         )
-        results = asyncio.run(mcp_search(query=search_query, page=page, page_size=page_size))
         # Use json module for more controlled serialization
         import json
 

src/basic_memory/config.py

@@ -234,6 +234,7 @@ def get_process_name():  # pragma: no cover
 # Global flag to track if logging has been set up
 _LOGGING_SETUP = False
 
+
 def setup_basic_memory_logging():  # pragma: no cover
     """Set up logging for basic-memory, ensuring it only happens once."""
     global _LOGGING_SETUP

src/basic_memory/mcp/prompts/continue_conversation.py

@@ -5,19 +5,19 @@
 """
 
 from textwrap import dedent
-from typing import Optional, Annotated
+from typing import Annotated, Optional
 
 from loguru import logger
 from pydantic import Field
 
-from basic_memory.mcp.prompts.utils import format_prompt_context, PromptContext, PromptContextItem
+from basic_memory.mcp.prompts.utils import PromptContext, PromptContextItem, format_prompt_context
 from basic_memory.mcp.server import mcp
 from basic_memory.mcp.tools.build_context import build_context
 from basic_memory.mcp.tools.recent_activity import recent_activity
 from basic_memory.mcp.tools.search import search_notes
 from basic_memory.schemas.base import TimeFrame
 from basic_memory.schemas.memory import GraphContext
-from basic_memory.schemas.search import SearchQuery, SearchItemType
+from basic_memory.schemas.search import SearchItemType
 
 
 @mcp.prompt(
@@ -48,7 +48,7 @@ async def continue_conversation(
     # If topic provided, search for it
     if topic:
         search_results = await search_notes(
-            SearchQuery(text=topic, after_date=timeframe, types=[SearchItemType.ENTITY])
+            query=topic, after_date=timeframe, entity_types=[SearchItemType.ENTITY]
         )
 
         # Build context from results

src/basic_memory/mcp/prompts/search.py

@@ -12,7 +12,7 @@
 from basic_memory.mcp.server import mcp
 from basic_memory.mcp.tools.search import search_notes as search_tool
 from basic_memory.schemas.base import TimeFrame
-from basic_memory.schemas.search import SearchQuery, SearchResponse
+from basic_memory.schemas.search import SearchResponse
 
 
 @mcp.prompt(
@@ -40,7 +40,7 @@ async def search_prompt(
     """
     logger.info(f"Searching knowledge base, query: {query}, timeframe: {timeframe}")
 
-    search_results = await search_tool(SearchQuery(text=query, after_date=timeframe))
+    search_results = await search_tool(query=query, after_date=timeframe)
     return format_search_results(query, search_results, timeframe)
 
 

src/basic_memory/mcp/tools/read_note.py

@@ -9,7 +9,6 @@
 from basic_memory.mcp.tools.search import search_notes
 from basic_memory.mcp.tools.utils import call_get
 from basic_memory.schemas.memory import memory_url_path
-from basic_memory.schemas.search import SearchQuery
 
 
 @mcp.tool(
@@ -63,7 +62,7 @@ async def read_note(identifier: str, page: int = 1, page_size: int = 10) -> str:
 
     # Fallback 1: Try title search via API
     logger.info(f"Search title for: {identifier}")
-    title_results = await search_notes(SearchQuery(title=identifier))
+    title_results = await search_notes(query=identifier, search_type="title")
 
     if title_results and title_results.results:
         result = title_results.results[0]  # Get the first/best match
@@ -87,7 +86,7 @@ async def read_note(identifier: str, page: int = 1, page_size: int = 10) -> str:
 
     # Fallback 2: Text search as a last resort
     logger.info(f"Title search failed, trying text search for: {identifier}")
-    text_results = await search_notes(SearchQuery(text=identifier))
+    text_results = await search_notes(query=identifier, search_type="text")
 
     # We didn't find a direct match, construct a helpful error message
     if not text_results or not text_results.results: