zrb 1.4.1__tar.gz → 1.4.3__tar.gz

{zrb-1.4.1 → zrb-1.4.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: zrb
-Version: 1.4.1
+Version: 1.4.3
 Summary: Your Automation Powerhouse
 Home-page: https://github.com/state-alchemists/zrb
 License: AGPL-3.0-or-later
@@ -13,16 +13,20 @@ Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Provides-Extra: all
+Provides-Extra: playwright
 Provides-Extra: rag
 Requires-Dist: autopep8 (>=2.0.4,<3.0.0)
 Requires-Dist: beautifulsoup4 (>=4.12.3,<5.0.0)
 Requires-Dist: black (>=24.10.0,<24.11.0)
-Requires-Dist: chromadb (>=0.5.20,<0.6.0) ; extra == "rag"
+Requires-Dist: chromadb (>=0.5.20,<0.6.0) ; extra == "rag" or extra == "all"
 Requires-Dist: fastapi[standard] (>=0.115.6,<0.116.0)
 Requires-Dist: fastembed (>=0.5.1,<0.6.0)
 Requires-Dist: isort (>=5.13.2,<5.14.0)
 Requires-Dist: libcst (>=1.5.0,<2.0.0)
-Requires-Dist: pdfplumber (>=0.11.4,<0.12.0) ; extra == "rag"
+Requires-Dist: openai (>=1.10.0,<2.0.0) ; extra == "rag" or extra == "all"
+Requires-Dist: pdfplumber (>=0.11.4,<0.12.0) ; extra == "rag" or extra == "all"
+Requires-Dist: playwright (>=1.43.0,<2.0.0) ; extra == "playwright" or extra == "all"
 Requires-Dist: psutil (>=6.1.1,<7.0.0)
 Requires-Dist: pydantic-ai (>=0.0.42,<0.0.43)
 Requires-Dist: python-dotenv (>=1.0.1,<2.0.0)
@@ -44,7 +48,6 @@ Description-Content-Type: text/markdown
 
 Zrb streamlines repetitive tasks, integrates with powerful LLMs, and lets you create custom automation workflows effortlessly. Whether you’re building CI/CD pipelines, code generators, or unique automation scripts, Zrb is designed to simplify and supercharge your workflow.
 
----
 
 ## 🚀 Why Zrb?
 
@@ -54,7 +57,6 @@ Zrb streamlines repetitive tasks, integrates with powerful LLMs, and lets you cr
 - **Developer-Friendly:** Quick to install and get started, with clear documentation and examples.
 - **Web Interface:** Run Zrb as a server to make tasks accessible even to non-technical team members.
 
----
 
 ## 🔥 Key Features
 
@@ -64,11 +66,8 @@ Zrb streamlines repetitive tasks, integrates with powerful LLMs, and lets you cr
 - **Flexible Input Handling:** Defaults, prompts, and command-line parameters to suit any workflow.
 - **Extensible & Open Source:** Contribute, customize, or extend Zrb to fit your unique needs.
 
----
 
-## 🛠️ Getting Started
-
-### Quick Installation
+# 🛠️ Installation
 
 Install Zrb via pip:
 
@@ -84,40 +83,48 @@ bash -c "$(curl -fsSL https://raw.githubusercontent.com/state-alchemists/zrb/mai
 
 ```
 
-### Your First Task
+# 🍲 Quick Start: Build Your First Automation Workflow
+
+Zrb empowers you to create custom automation tasks using Python. This guide shows you how to define two simple tasks: one to generate a PlantUML script from your source code and another to convert that script into a PNG image.
+
+## 1. Create Your Task Definition File
 
-Create a file at `/home/<your-user-name>/zrb_init.py` with the following content:
+Place a file named `zrb_init.py` in a directory that's accessible from your projects. Zrb will automatically search for this file by starting in your current directory and then moving upward (i.e., checking parent directories) until it finds one. This means if you place your `zrb_init.py` in your home directory (e.g., `/home/<your-user-name>/zrb_init.py`), the tasks defined there will be available for any project.
 
+Add the following content to your zrb_init.py:
 
 ```python
 import os
-from zrb import cli, llm_config, LLMTask, CmdTask, StrInput, Group
-from zrb.builtin.llm.tool.file import read_all_files, write_text_file
+from zrb import cli, LLMTask, CmdTask, StrInput, Group
+from zrb.builtin.llm.tool.file import (
+    read_text_file, list_files, write_text_file
+)
 
 CURRENT_DIR = os.getcwd()
 
-# Make UML group
+# Create a group for UML-related tasks
 uml_group = cli.add_group(Group(name="uml", description="UML related tasks"))
 
-# Generate UML script
+# Task 1: Generate a PlantUML script from your source code
 make_uml_script = uml_group.add_task(
     LLMTask(
         name="make-script",
         description="Creating plantuml diagram based on source code in current directory",
         input=StrInput(name="diagram", default="state diagram"),
         message=(
-            f"Read source code in {CURRENT_DIR}, "
+            f"Read all necessary files in {CURRENT_DIR}, "
             "make a {ctx.input.diagram} in plantuml format. "
             f"Write the script into {CURRENT_DIR}/{{ctx.input.diagram}}.uml"
         ),
         tools=[
-            read_all_files,
+            list_files,
+            read_text_file,
             write_text_file,
         ],
     )
 )
 
-# Defining a Cmd Task to transform Plantuml script into a png image.
+# Task 2: Convert the PlantUML script into a PNG image
 make_uml_image = uml_group.add_task(
     CmdTask(
         name="make-image",
@@ -128,62 +135,75 @@ make_uml_image = uml_group.add_task(
     )
 )
 
-# Making sure that make_png has make_uml as its dependency.
+# Set up the dependency: the image task runs after the script is created
 make_uml_script >> make_uml_image
 ```
 
-You have just define two automation tasks.
+**What This Does**
 
-The first one use LLM to read files in your current directory and create a `PlantUML script` on that directory.
+- **Task 1 – make-script**:
 
-The second task turn the PlantUML script into a `*.png` file. The second task depends on the first task and both of them are located under the same group.
+    Uses an LLM to read all files in your current directory and generate a PlantUML script (e.g., `state diagram.uml`).
 
-You can run the tasks by invoking `zrb uml make-script` or `zrb uml make-image` respectively.
+- **Task 2 – make-image**:
 
-When you run zrb, it automatically searches for a file named `zrb_init.py` starting from your current directory and moving upward through its parent directories. This design lets you set up common automation tasks in a central location—like placing a `zrb_init.py` in your home directory (`/home/<your-user>/zrb_init.py`)—so that your tasks are available across all your projects.
+    Executes a command that converts the PlantUML script into a PNG image (e.g., `state diagram.png`). This task will run only after the script has been generated.
 
-Now, go to your project and create a state diagram:
 
-```bash
-git clone git@github.com:jjinux/gotetris.git
-cd gotetris
-zrb uml make-image --diagram "state diagram"
-```
+## 2. Run Your Tasks
 
-You can also invoke the task without specifying parameter.
+After setting up your tasks, you can execute them from any project. For example:
 
-```bash
-zrb uml make-image
-```
+- Clone/Create a Project:
 
-Once you do so, Zrb will ask you to provide the diagram type.
+    ```bash
+    git clone git@github.com:jjinux/gotetris.git
+    cd gotetris
+    ```
 
-```
-diagram [state diagram]:
-```
+- Create a state diagram:
+
+    ```bash
+    zrb uml make-image --diagram "state diagram"
+    ```
+
+- Or use the interactive mode:
 
-You can just press enter if you want to use the default value (i.e., in this case `state diagram`).
+    ```bash
+    zrb uml make-image
+    ```
 
-Finally, you can also serve the tasks via a Web UI interface by invoking the following command:
+    Zrb will prompt:
+
+    ```bash
+    diagram [state diagram]:
+    ```
+
+    Press **Enter** to use the default value
+
+![State Diagram](https://raw.githubusercontent.com/state-alchemists/zrb/main/_images/state-diagram.png)
+
+
+## 3. Try Out the Web UI
+
+You can also serve your tasks through a user-friendly web interface:
 
 ```bash
 zrb server start
 ```
 
-You will have a nice web interface running on `http://localhost:12123`
+Then open your browser and visit `http://localhost:21213`
 
 ![Zrb Web UI](https://raw.githubusercontent.com/state-alchemists/zrb/main/_images/zrb-web-ui.png)
 
-Now, let's see how things work in detail. First, Zrb generates a `state diagram.uml` in your current directory, it then transform the UML script into a PNG image `state diagram.png`.
-
-![State Diagram](https://raw.githubusercontent.com/state-alchemists/zrb/main/_images/state-diagram.png)
-
 
 # 🎥 Demo & Documentation
 
 - **Step by step guide:** [Getting started with Zrb](https://github.com/state-alchemists/zrb/blob/main/docs/recipes/getting-started/README.md).
 - **Full documentation:** [Zrb Documentation](https://github.com/state-alchemists/zrb/blob/main/docs/README.md)
-- **Video demo:** [![Video Title](https://img.youtube.com/vi/W7dgk96l__o/0.jpg)](https://www.youtube.com/watch?v=W7dgk96l__o)
+- **Video demo:**
+
+    [![Video Title](https://img.youtube.com/vi/W7dgk96l__o/0.jpg)](https://www.youtube.com/watch?v=W7dgk96l__o)
 
 
 # 🤝 Join the Community

{zrb-1.4.1 → zrb-1.4.3}/README.md

@@ -9,7 +9,6 @@
 
 Zrb streamlines repetitive tasks, integrates with powerful LLMs, and lets you create custom automation workflows effortlessly. Whether you’re building CI/CD pipelines, code generators, or unique automation scripts, Zrb is designed to simplify and supercharge your workflow.
 
----
 
 ## 🚀 Why Zrb?
 
@@ -19,7 +18,6 @@ Zrb streamlines repetitive tasks, integrates with powerful LLMs, and lets you cr
 - **Developer-Friendly:** Quick to install and get started, with clear documentation and examples.
 - **Web Interface:** Run Zrb as a server to make tasks accessible even to non-technical team members.
 
----
 
 ## 🔥 Key Features
 
@@ -29,11 +27,8 @@ Zrb streamlines repetitive tasks, integrates with powerful LLMs, and lets you cr
 - **Flexible Input Handling:** Defaults, prompts, and command-line parameters to suit any workflow.
 - **Extensible & Open Source:** Contribute, customize, or extend Zrb to fit your unique needs.
 
----
 
-## 🛠️ Getting Started
-
-### Quick Installation
+# 🛠️ Installation
 
 Install Zrb via pip:
 
@@ -49,40 +44,48 @@ bash -c "$(curl -fsSL https://raw.githubusercontent.com/state-alchemists/zrb/mai
 
 ```
 
-### Your First Task
+# 🍲 Quick Start: Build Your First Automation Workflow
+
+Zrb empowers you to create custom automation tasks using Python. This guide shows you how to define two simple tasks: one to generate a PlantUML script from your source code and another to convert that script into a PNG image.
+
+## 1. Create Your Task Definition File
 
-Create a file at `/home/<your-user-name>/zrb_init.py` with the following content:
+Place a file named `zrb_init.py` in a directory that's accessible from your projects. Zrb will automatically search for this file by starting in your current directory and then moving upward (i.e., checking parent directories) until it finds one. This means if you place your `zrb_init.py` in your home directory (e.g., `/home/<your-user-name>/zrb_init.py`), the tasks defined there will be available for any project.
 
+Add the following content to your zrb_init.py:
 
 ```python
 import os
-from zrb import cli, llm_config, LLMTask, CmdTask, StrInput, Group
-from zrb.builtin.llm.tool.file import read_all_files, write_text_file
+from zrb import cli, LLMTask, CmdTask, StrInput, Group
+from zrb.builtin.llm.tool.file import (
+    read_text_file, list_files, write_text_file
+)
 
 CURRENT_DIR = os.getcwd()
 
-# Make UML group
+# Create a group for UML-related tasks
 uml_group = cli.add_group(Group(name="uml", description="UML related tasks"))
 
-# Generate UML script
+# Task 1: Generate a PlantUML script from your source code
 make_uml_script = uml_group.add_task(
     LLMTask(
         name="make-script",
         description="Creating plantuml diagram based on source code in current directory",
         input=StrInput(name="diagram", default="state diagram"),
         message=(
-            f"Read source code in {CURRENT_DIR}, "
+            f"Read all necessary files in {CURRENT_DIR}, "
             "make a {ctx.input.diagram} in plantuml format. "
             f"Write the script into {CURRENT_DIR}/{{ctx.input.diagram}}.uml"
         ),
         tools=[
-            read_all_files,
+            list_files,
+            read_text_file,
             write_text_file,
         ],
     )
 )
 
-# Defining a Cmd Task to transform Plantuml script into a png image.
+# Task 2: Convert the PlantUML script into a PNG image
 make_uml_image = uml_group.add_task(
     CmdTask(
         name="make-image",
@@ -93,62 +96,75 @@ make_uml_image = uml_group.add_task(
     )
 )
 
-# Making sure that make_png has make_uml as its dependency.
+# Set up the dependency: the image task runs after the script is created
 make_uml_script >> make_uml_image
 ```
 
-You have just define two automation tasks.
+**What This Does**
 
-The first one use LLM to read files in your current directory and create a `PlantUML script` on that directory.
+- **Task 1 – make-script**:
 
-The second task turn the PlantUML script into a `*.png` file. The second task depends on the first task and both of them are located under the same group.
+    Uses an LLM to read all files in your current directory and generate a PlantUML script (e.g., `state diagram.uml`).
 
-You can run the tasks by invoking `zrb uml make-script` or `zrb uml make-image` respectively.
+- **Task 2 – make-image**:
 
-When you run zrb, it automatically searches for a file named `zrb_init.py` starting from your current directory and moving upward through its parent directories. This design lets you set up common automation tasks in a central location—like placing a `zrb_init.py` in your home directory (`/home/<your-user>/zrb_init.py`)—so that your tasks are available across all your projects.
+    Executes a command that converts the PlantUML script into a PNG image (e.g., `state diagram.png`). This task will run only after the script has been generated.
 
-Now, go to your project and create a state diagram:
 
-```bash
-git clone git@github.com:jjinux/gotetris.git
-cd gotetris
-zrb uml make-image --diagram "state diagram"
-```
+## 2. Run Your Tasks
 
-You can also invoke the task without specifying parameter.
+After setting up your tasks, you can execute them from any project. For example:
 
-```bash
-zrb uml make-image
-```
+- Clone/Create a Project:
 
-Once you do so, Zrb will ask you to provide the diagram type.
+    ```bash
+    git clone git@github.com:jjinux/gotetris.git
+    cd gotetris
+    ```
 
-```
-diagram [state diagram]:
-```
+- Create a state diagram:
+
+    ```bash
+    zrb uml make-image --diagram "state diagram"
+    ```
+
+- Or use the interactive mode:
 
-You can just press enter if you want to use the default value (i.e., in this case `state diagram`).
+    ```bash
+    zrb uml make-image
+    ```
 
-Finally, you can also serve the tasks via a Web UI interface by invoking the following command:
+    Zrb will prompt:
+
+    ```bash
+    diagram [state diagram]:
+    ```
+
+    Press **Enter** to use the default value
+
+![State Diagram](https://raw.githubusercontent.com/state-alchemists/zrb/main/_images/state-diagram.png)
+
+
+## 3. Try Out the Web UI
+
+You can also serve your tasks through a user-friendly web interface:
 
 ```bash
 zrb server start
 ```
 
-You will have a nice web interface running on `http://localhost:12123`
+Then open your browser and visit `http://localhost:21213`
 
 ![Zrb Web UI](https://raw.githubusercontent.com/state-alchemists/zrb/main/_images/zrb-web-ui.png)
 
-Now, let's see how things work in detail. First, Zrb generates a `state diagram.uml` in your current directory, it then transform the UML script into a PNG image `state diagram.png`.
-
-![State Diagram](https://raw.githubusercontent.com/state-alchemists/zrb/main/_images/state-diagram.png)
-
 
 # 🎥 Demo & Documentation
 
 - **Step by step guide:** [Getting started with Zrb](https://github.com/state-alchemists/zrb/blob/main/docs/recipes/getting-started/README.md).
 - **Full documentation:** [Zrb Documentation](https://github.com/state-alchemists/zrb/blob/main/docs/README.md)
-- **Video demo:** [![Video Title](https://img.youtube.com/vi/W7dgk96l__o/0.jpg)](https://www.youtube.com/watch?v=W7dgk96l__o)
+- **Video demo:**
+
+    [![Video Title](https://img.youtube.com/vi/W7dgk96l__o/0.jpg)](https://www.youtube.com/watch?v=W7dgk96l__o)
 
 
 # 🤝 Join the Community

{zrb-1.4.1 → zrb-1.4.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "zrb"
-version = "1.4.1"
+version = "1.4.3"
 description = "Your Automation Powerhouse"
 authors = ["Go Frendi Gunawan <gofrendiasgard@gmail.com>"]
 license = "AGPL-3.0-or-later"
@@ -52,11 +52,15 @@ python-jose = {extras = ["cryptography"], version = "^3.4.0"}
 ulid-py = "^1.1.0"
 pydantic-ai = "^0.0.42"
 fastembed = "^0.5.1"
+openai = {version = "^1.10.0", optional = true}
+playwright = {version = "^1.43.0", optional = true}
 psutil = "^6.1.1"
 
 [tool.poetry.extras]
 # poetry install -E rag
-rag = ["chromadb", "pdfplumber"]
+rag = ["chromadb", "pdfplumber", "openai"]
+playwright = ["playwright"]
+all = ["chromadb", "pdfplumber", "openai", "playwright"]
 
 [tool.poetry.group.dev.dependencies]
 alembic = "^1.14.0" # FastApp dependencies

{zrb-1.4.1 → zrb-1.4.3}/src/zrb/builtin/llm/llm_chat.py

@@ -22,7 +22,6 @@ from zrb.config import (
     LLM_ALLOW_ACCESS_LOCAL_FILE,
     LLM_ALLOW_ACCESS_SHELL,
     LLM_HISTORY_DIR,
-    LLM_SYSTEM_PROMPT,
     SERP_API_KEY,
 )
 from zrb.context.any_shared_context import AnySharedContext
@@ -119,7 +118,7 @@ llm_chat: LLMTask = llm_group.add_task(
                 "system-prompt",
                 description="System prompt",
                 prompt="System prompt",
-                default=LLM_SYSTEM_PROMPT,
+                default="",
                 allow_positional_parsing=False,
                 always_prompt=False,
             ),
@@ -141,17 +140,19 @@ llm_chat: LLMTask = llm_group.add_task(
                 always_prompt=False,
             ),
         ],
-        model=lambda ctx: None if ctx.input.model == "" else ctx.input.model,
+        model=lambda ctx: None if ctx.input.model.strip() == "" else ctx.input.model,
         model_base_url=lambda ctx: (
-            None if ctx.input.base_url == "" else ctx.input.base_url
+            None if ctx.input.base_url.strip() == "" else ctx.input.base_url
         ),
         model_api_key=lambda ctx: (
-            None if ctx.input.api_key == "" else ctx.input.api_key
+            None if ctx.input.api_key.strip() == "" else ctx.input.api_key
         ),
         conversation_history_reader=_read_chat_conversation,
         conversation_history_writer=_write_chat_conversation,
         description="Chat with LLM",
-        system_prompt="{ctx.input.system_prompt}",
+        system_prompt=lambda ctx: (
+            None if ctx.input.system_prompt.strip() == "" else ctx.input.system_prompt
+        ),
         message="{ctx.input.message}",
         retries=0,
     ),

{zrb-1.4.1 → zrb-1.4.3}/src/zrb/builtin/llm/tool/api.py

@@ -5,7 +5,7 @@ from typing import Annotated, Literal
 def get_current_location() -> (
     Annotated[str, "JSON string representing latitude and longitude"]
 ):  # noqa
-    """Get the user's current location."""
+    """Get the user's current location. This function take no argument."""
     import requests
 
     return json.dumps(requests.get("http://ip-api.com/json?fields=lat,lon").json())

{zrb-1.4.1 → zrb-1.4.3}/src/zrb/builtin/llm/tool/file.py

@@ -93,6 +93,7 @@ def list_files(
     """List all files in a directory that match any of the included glob patterns
     and do not reside in any directory matching an excluded pattern.
     Patterns are evaluated using glob-style matching.
+    included_patterns and excluded_patterns already has sane default values.
     """
     all_files: list[str] = []
     for root, dirs, files in os.walk(directory):
@@ -127,12 +128,12 @@ def _should_exclude(full_path: str, excluded_patterns: list[str]) -> bool:
 
 
 def read_text_file(file: str) -> str:
-    """Read a text file"""
+    """Read a text file and return a string containing the file content."""
     return read_file(os.path.abspath(file))
 
 
 def write_text_file(file: str, content: str):
-    """Write a text file"""
+    """Write content to a text file"""
     return write_file(os.path.abspath(file), content)
 
 
@@ -144,6 +145,7 @@ def read_all_files(
     """Read all files in a directory that match any of the included glob patterns
     and do not match any of the excluded glob patterns.
     Patterns are evaluated using glob-style matching.
+    included_patterns and excluded_patterns already has sane default values.
     """
     files = list_files(directory, included_patterns, excluded_patterns)
     for index, file in enumerate(files):

{zrb-1.4.1 → zrb-1.4.3}/src/zrb/builtin/llm/tool/rag.py

@@ -9,6 +9,8 @@ import ulid
 
 from zrb.config import (
     RAG_CHUNK_SIZE,
+    RAG_EMBEDDING_API_KEY,
+    RAG_EMBEDDING_BASE_URL,
     RAG_EMBEDDING_MODEL,
     RAG_MAX_RESULT_COUNT,
     RAG_OVERLAP,
@@ -35,24 +37,34 @@ def create_rag_from_directory(
     tool_name: str,
     tool_description: str,
     document_dir_path: str = "./documents",
-    model: str = RAG_EMBEDDING_MODEL,
     vector_db_path: str = "./chroma",
     vector_db_collection: str = "documents",
     chunk_size: int = RAG_CHUNK_SIZE,
     overlap: int = RAG_OVERLAP,
     max_result_count: int = RAG_MAX_RESULT_COUNT,
     file_reader: list[RAGFileReader] = [],
+    openai_api_key: str = RAG_EMBEDDING_API_KEY,
+    openai_base_url: str = RAG_EMBEDDING_BASE_URL,
+    openai_embedding_model: str = RAG_EMBEDDING_MODEL,
 ):
     async def retrieve(query: str) -> str:
         from chromadb import PersistentClient
         from chromadb.config import Settings
-        from fastembed import TextEmbedding
-
-        embedding_model = TextEmbedding(model_name=model)
-        client = PersistentClient(
+        from openai import OpenAI
+
+        # Initialize OpenAI client with custom URL if provided
+        client_args = {}
+        if openai_api_key:
+            client_args["api_key"] = openai_api_key
+        if openai_base_url:
+            client_args["base_url"] = openai_base_url
+        # Initialize OpenAI client for embeddings
+        openai_client = OpenAI(**client_args)
+        # Initialize ChromaDB client
+        chroma_client = PersistentClient(
             path=vector_db_path, settings=Settings(allow_reset=True)
         )
-        collection = client.get_or_create_collection(vector_db_collection)
+        collection = chroma_client.get_or_create_collection(vector_db_collection)
         # Track file changes using a hash-based approach
         hash_file_path = os.path.join(vector_db_path, "file_hashes.json")
         previous_hashes = _load_hashes(hash_file_path)
@@ -89,8 +101,11 @@ def create_rag_from_directory(
                                 ),
                                 file=sys.stderr,
                             )
-                            embedding_result = list(embedding_model.embed([chunk]))
-                            vector = embedding_result[0]
+                            # Get embeddings using OpenAI
+                            embedding_response = openai_client.embeddings.create(
+                                input=chunk, model=openai_embedding_model
+                            )
+                            vector = embedding_response.data[0].embedding
                             collection.upsert(
                                 ids=[chunk_id],
                                 embeddings=[vector],
@@ -113,8 +128,11 @@ def create_rag_from_directory(
             )
         # Vectorize query and get related document chunks
         print(stylize_faint("Vectorizing query"), file=sys.stderr)
-        embedding_result = list(embedding_model.embed([query]))
-        query_vector = embedding_result[0]
+        # Get embeddings using OpenAI
+        embedding_response = openai_client.embeddings.create(
+            input=query, model=openai_embedding_model
+        )
+        query_vector = embedding_response.data[0].embedding
         print(stylize_faint("Searching documents"), file=sys.stderr)
         results = collection.query(
             query_embeddings=query_vector,