chatterer 0.1.12__py3-none-any.whl → 0.1.14__py3-none-any.whl

chatterer-0.1.14.dist-info/METADATA

@@ -0,0 +1,387 @@
+Metadata-Version: 2.4
+Name: chatterer
+Version: 0.1.14
+Summary: The highest-level interface for various LLM APIs.
+Requires-Python: >=3.12
+Description-Content-Type: text/markdown
+Requires-Dist: instructor>=1.7.2
+Requires-Dist: langchain>=0.3.19
+Requires-Dist: langchain-openai>=0.3.11
+Requires-Dist: pillow>=11.1.0
+Requires-Dist: regex>=2024.11.6
+Provides-Extra: dev
+Requires-Dist: neo4j-extension>=0.1.14; extra == "dev"
+Requires-Dist: colorama>=0.4.6; extra == "dev"
+Requires-Dist: ipykernel>=6.29.5; extra == "dev"
+Provides-Extra: conversion
+Requires-Dist: youtube-transcript-api>=1.0.3; extra == "conversion"
+Requires-Dist: chatterer[browser]; extra == "conversion"
+Requires-Dist: chatterer[pdf]; extra == "conversion"
+Requires-Dist: chatterer[markdown]; extra == "conversion"
+Provides-Extra: browser
+Requires-Dist: playwright>=1.50.0; extra == "browser"
+Provides-Extra: pdf
+Requires-Dist: pymupdf>=1.25.4; extra == "pdf"
+Requires-Dist: pypdf>=5.4.0; extra == "pdf"
+Provides-Extra: markdown
+Requires-Dist: markitdown[all]>=0.1.1; extra == "markdown"
+Requires-Dist: markdownify>=1.1.0; extra == "markdown"
+Requires-Dist: commonmark>=0.9.1; extra == "markdown"
+Requires-Dist: mistune>=3.1.3; extra == "markdown"
+Provides-Extra: langchain
+Requires-Dist: chatterer[langchain-providers]; extra == "langchain"
+Requires-Dist: langchain-experimental>=0.3.4; extra == "langchain"
+Provides-Extra: langchain-providers
+Requires-Dist: langchain-anthropic>=0.3.10; extra == "langchain-providers"
+Requires-Dist: langchain-google-genai>=2.1.1; extra == "langchain-providers"
+Requires-Dist: langchain-ollama>=0.3.0; extra == "langchain-providers"
+Provides-Extra: all
+Requires-Dist: chatterer[dev]; extra == "all"
+Requires-Dist: chatterer[langchain]; extra == "all"
+Requires-Dist: chatterer[conversion]; extra == "all"
+
+# Chatterer
+
+**Simplified, Structured AI Assistant Framework**
+
+`chatterer` is a Python library designed as a type-safe LangChain wrapper for interacting with various language models (OpenAI, Anthropic, Google Gemini, Ollama, etc.). It supports structured outputs via Pydantic models, plain text responses, asynchronous calls, image description, code execution, and an interactive shell.
+
+The structured reasoning in `chatterer` is inspired by the [Atom-of-Thought](https://github.com/qixucen/atom) pipeline.
+
+---
+
+## Quick Install
+
+```bash
+pip install chatterer
+```
+
+---
+
+## Quickstart Example
+
+Generate text quickly using OpenAI. 
+Messages can be input as plain strings or structured lists:
+
+```python
+from chatterer import Chatterer, HumanMessage, AIMessage, SystemMessage
+
+# Initialize the Chatterer with `openai`, `anthropic`, `google`, or `ollama` models
+chatterer: Chatterer = Chatterer.openai("gpt-4.1")
+
+# Get direct response as str
+response: str = chatterer("What is the meaning of life?")
+# response = chatterer([{ "role": "user", "content": "What is the meaning of life?" }])
+# response = chatterer([("user", "What is the meaning of life?")])
+# response = chatterer([HumanMessage("What is the meaning of life?")])
+print(response)
+```
+
+Image & text content can be sent as together:
+
+```python
+from chatterer import Base64Image, HumanMessage
+
+# Load an image from a file or URL, resulting in a None or Base64Image object
+image = Base64Image.from_url_or_path("example.jpg")
+# image = Base64Image.from_url_or_path("https://example.com/image.jpg")
+assert image is not None, "Failed to load image"
+
+# Alternatively, load an image from bytes
+# with open("example.jpg", "rb") as f:
+#     image = Base64Image.from_bytes(f.read(), ext="jpeg")
+
+message = HumanMessage(["Describe the image", image.data_uri_content])
+response: str = chatterer([message])
+print(response)
+```
+
+---
+
+## Structured Output with Pydantic
+
+Define a Pydantic model and get typed responses:
+
+```python
+from pydantic import BaseModel
+
+class AnswerModel(BaseModel):
+    question: str
+    answer: str
+
+# Call with response_model
+response: AnswerModel = chatterer("What's the capital of France?", response_model=AnswerModel)
+print(response.question, response.answer)
+```
+
+---
+
+## Async Example
+
+Use asynchronous generation for non-blocking operations:
+
+```python
+import asyncio
+
+async def main():
+    response = await chatterer.agenerate("Explain async in Python briefly.")
+    print(response)
+
+asyncio.run(main())
+```
+
+---
+
+## Streaming Structured Outputs
+
+Stream structured responses in real-time:
+
+```python
+from pydantic import BaseModel
+
+class AnswerModel(BaseModel):
+    text: str
+
+chatterer = Chatterer.openai()
+for chunk in chatterer.generate_pydantic_stream(AnswerModel, "Tell me a story"):
+    print(chunk.text)
+```
+
+Asynchronous version:
+```python
+import asyncio
+
+async def main():
+    async for chunk in chatterer.agenerate_pydantic_stream(AnswerModel, "Tell me a story"):
+        print(chunk.text)
+
+asyncio.run(main())
+```
+
+---
+
+## Image Description
+
+Generate descriptions for images using the language model:
+
+```python
+description = chatterer.describe_image("https://example.com/image.jpg")
+print(description)
+
+# Customize the instruction
+description = chatterer.describe_image("https://example.com/image.jpg", instruction="Describe the main objects in the image.")
+```
+
+An asynchronous version is also available:
+
+```python
+async def main():
+    description = await chatterer.adescribe_image("https://example.com/image.jpg")
+    print(description)
+
+asyncio.run(main())
+```
+
+---
+
+## Code Execution
+
+Generate and execute Python code dynamically:
+
+```python
+result = chatterer.invoke_code_execution("Write a function to calculate factorial.")
+print(result.code)
+print(result.output)
+```
+
+An asynchronous version exists as well:
+
+```python
+async def main():
+    result = await chatterer.ainvoke_code_execution("Write a function to calculate factorial.")
+    print(result.output)
+
+asyncio.run(main())
+```
+
+---
+
+## Webpage to Markdown
+
+Convert webpages to Markdown, optionally filtering content with the language model:
+
+```python
+from chatterer.tools.webpage_to_markdown import PlayWrightBot
+
+with PlayWrightBot() as bot:
+    # Basic conversion
+    markdown = bot.url_to_md("https://example.com")
+    print(markdown)
+
+    # With LLM filtering and image descriptions
+    filtered_md = bot.url_to_md_with_llm("https://example.com", describe_images=True)
+    print(filtered_md)
+```
+
+Asynchronous version:
+```python
+import asyncio
+
+async def main():
+    async with PlayWrightBot() as bot:
+        markdown = await bot.aurl_to_md_with_llm("https://example.com")
+        print(markdown)
+
+asyncio.run(main())
+```
+
+Extract specific elements:
+```python
+with PlayWrightBot() as bot:
+    headings = bot.select_and_extract("https://example.com", "h2")
+    print(headings)
+```
+
+---
+
+## Citation Chunking
+
+Chunk documents into semantic sections with citations:
+
+```python
+from chatterer import Chatterer
+from chatterer.tools import citation_chunker
+
+chatterer = Chatterer.openai()
+document = "Long text about quantum computing..."
+chunks = citation_chunker(document, chatterer, global_coverage_threshold=0.9)
+for chunk in chunks:
+    print(f"Subject: {chunk.name}")
+    for source, matches in chunk.references.items():
+        print(f"  Source: {source}, Matches: {matches}")
+```
+
+---
+
+## Interactive Shell
+
+Engage in a conversational AI session with code execution support:
+
+```python
+from chatterer import interactive_shell
+
+interactive_shell()
+```
+
+This launches an interactive session where you can chat with the AI and execute code snippets. Type `quit` or `exit` to end the session.
+
+---
+
+## Atom-of-Thought Pipeline (AoT)
+
+`AoTPipeline` provides structured reasoning inspired by the [Atom-of-Thought](https://github.com/qixucen/atom) approach. It decomposes complex questions recursively, generates answers, and combines them via an ensemble process.
+
+### AoT Usage Example
+
+```python
+from chatterer import Chatterer
+from chatterer.strategies import AoTStrategy, AoTPipeline
+
+pipeline = AoTPipeline(chatterer=Chatterer.openai(), max_depth=2)
+strategy = AoTStrategy(pipeline=pipeline)
+
+question = "What would Newton discover if hit by an apple falling from 100 meters?"
+answer = strategy.invoke(question)
+print(answer)
+
+# Generate and inspect reasoning graph
+graph = strategy.get_reasoning_graph()
+print(f"Graph: {len(graph.nodes)} nodes, {len(graph.relationships)} relationships")
+```
+
+**Note**: The AoT pipeline includes an optional feature to generate a reasoning graph, which can be stored in Neo4j for visualization and analysis. Install `neo4j_extension` and set up a Neo4j instance to use this feature:
+
+```python
+from neo4j_extension import Neo4jConnection
+with Neo4jConnection() as conn:
+    conn.upsert_graph(graph)
+```
+
+---
+
+## Supported Models
+
+Chatterer supports multiple language models, easily initialized as follows:
+
+- **OpenAI**
+- **Anthropic**
+- **Google Gemini**
+- **Ollama** (local models)
+
+```python
+openai_chatterer = Chatterer.openai("gpt-4o-mini")
+anthropic_chatterer = Chatterer.anthropic("claude-3-7-sonnet-20250219")
+gemini_chatterer = Chatterer.google("gemini-2.0-flash")
+ollama_chatterer = Chatterer.ollama("deepseek-r1:1.5b")
+```
+
+---
+
+## Advanced Features
+
+- **Streaming Responses**: Use `generate_stream` or `agenerate_stream` for real-time output.
+- **Streaming Structured Outputs**: Stream Pydantic-typed responses with `generate_pydantic_stream` or `agenerate_pydantic_stream`.
+- **Async/Await Support**: All methods have asynchronous counterparts (e.g., `agenerate`, `adescribe_image`).
+- **Structured Outputs**: Leverage Pydantic models for typed responses.
+- **Image Description**: Generate descriptions for images with `describe_image`.
+- **Code Execution**: Dynamically generate and execute Python code with `invoke_code_execution`.
+- **Webpage to Markdown**: Convert webpages to Markdown with `PlayWrightBot`, including JavaScript rendering, element extraction, and LLM-based content filtering.
+- **Citation Chunking**: Semantically chunk documents and extract citations with `citation_chunker`, including coverage analysis.
+- **Interactive Shell**: Use `interactive_shell` for conversational AI with code execution.
+- **Token Counting**: Retrieve input/output token counts with `get_num_tokens_from_message`.
+- **Utilities**: Tools for content processing (e.g., `html_to_markdown`, `pdf_to_text`, `get_youtube_video_subtitle`, `citation_chunker`) are available in the `tools` module.
+
+```python
+# Example: Convert PDF to text
+from chatterer.tools import pdf_to_text
+text = pdf_to_text("example.pdf")
+print(text)
+
+# Example: Get YouTube subtitles
+from chatterer.tools import get_youtube_video_subtitle
+subtitles = get_youtube_video_subtitle("https://www.youtube.com/watch?v=example")
+print(subtitles)
+
+# Example: Get token counts
+from chatterer.messages import HumanMessage
+msg = HumanMessage(content="Hello, world!")
+tokens = chatterer.get_num_tokens_from_message(msg)
+if tokens:
+    input_tokens, output_tokens = tokens
+    print(f"Input: {input_tokens}, Output: {output_tokens}")
+```
+
+---
+
+## Logging
+
+Enable debugging with basic logging:
+
+```python
+import logging
+logging.basicConfig(level=logging.DEBUG)
+```
+
+The AoT pipeline uses a custom color-coded logger for detailed step-by-step output.
+
+---
+
+## Contributing
+
+We welcome contributions! Feel free to open an issue or submit a pull request on the repository.
+
+---
+
+## License
+
+MIT License

chatterer-0.1.14.dist-info/RECORD

@@ -0,0 +1,34 @@
+chatterer/__init__.py,sha256=Mxt_qGUyLEbbJ3VU-qr1EBgE2tSNFvqxiL-1QcXLu2c,2252
+chatterer/interactive.py,sha256=4_XKcdpRNLWEbEXwszJG-sECGun7_U1tEhd30brUPZM,16784
+chatterer/language_model.py,sha256=L3mVbbtiRoUx1UfJy0gCJ4laf03nyfDiDQtbEcB5Nv0,17594
+chatterer/messages.py,sha256=SIvG9hMHaPG4AFadeRbj5yPnMq2J06fHA4D8jrkz4kQ,458
+chatterer/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+chatterer/common_types/__init__.py,sha256=Ais0kqzQJj4sED24xdv5azIxHkkj0vXWb4LC41dFkBQ,355
+chatterer/common_types/io.py,sha256=GBZVhcRRCNZTAC1OPKAwW4s3EiyYRmNZ0ZWD1DxgMzs,798
+chatterer/strategies/__init__.py,sha256=oroDpp5ppWralCER5wnf8fgX77dqLAPF0ogmRtRzQfU,208
+chatterer/strategies/atom_of_thoughts.py,sha256=g801rY7k5UeNOq2-XRqB4h8sXjikVNzN57G6OtiUH00,39897
+chatterer/strategies/base.py,sha256=rqOzTo6S2eQ3A_F9aBXCmVoLM1eT6F2VzZ2Dof330Tk,413
+chatterer/tools/__init__.py,sha256=cOFo-Aj2xXK_7IvWYRdg6uNomaT9xuSa_mwk2Y0l_AM,1428
+chatterer/tools/caption_markdown_images.py,sha256=PfvHvr7x0XRLKlujALvOfEB3pQmjzlYFbcJqw2NHgZs,15008
+chatterer/tools/convert_pdf_to_markdown.py,sha256=prvgMvR3IPWopSVKfSIQRtnJNGnWUxLCwqXEgc6gcDA,14688
+chatterer/tools/convert_to_text.py,sha256=fMxUt7F6iMLf5tyZNNjhCxyW1zk7a6lTkDVe8YNS6XU,14996
+chatterer/tools/upstage_document_parser.py,sha256=xiXDUA95U0tAtxzVWcHHBHD72_TVmch-dg_QPmGNOGw,32398
+chatterer/tools/webpage_to_markdown.py,sha256=E9ZTg7fC6Z0dzGuYEsokRyzb8WVhiWdGFlQdiNSrBQ0,31222
+chatterer/tools/youtube.py,sha256=vrsf6pK6D6UBQQCiNUkE5Xe0VQ3SadoEnyAzGwhElX4,5935
+chatterer/tools/citation_chunking/__init__.py,sha256=DyLMGG4dVgSnGIdaSHcBNDz09iXflcKuJCtg4W0JTVo,79
+chatterer/tools/citation_chunking/chunks.py,sha256=_Sxzfbud8XTOHdHdQmKwm4-byES1cD1V6C28DgtS1BA,2120
+chatterer/tools/citation_chunking/citation_chunker.py,sha256=Aye1BqUCa4u_CsTZoqCe72pJA8C_y2U5UR7cNoNpeo4,4776
+chatterer/tools/citation_chunking/citations.py,sha256=BWhSwzZccvu0Db-OxEbsuEGEz-Dh_tXo8HRx1y2XUHg,12308
+chatterer/tools/citation_chunking/prompt.py,sha256=so-8uFQ5b2Zq2V5Brfxd76bEnKYkHovYsohAnbxWEnY,7557
+chatterer/tools/citation_chunking/reference.py,sha256=m47XYaB5uFff_x_k7US9hNr-SpZjKnl-GuzsGaQzcZo,893
+chatterer/tools/citation_chunking/utils.py,sha256=Xytm9lMrS783Po1qWAdEJ8q7Q3l2UMzwHd9EkYTRiwk,6210
+chatterer/utils/__init__.py,sha256=wA_nd1si3MbypA4biQRa3MYBqkHwTSCioYgaHce42Ls,412
+chatterer/utils/base64_image.py,sha256=w5SQoHVPDEQtFSUhV1l9GKg-IQy1WK5nyjv9k0upow8,10839
+chatterer/utils/bytesio.py,sha256=QabdJCZsabPaiYVfJcdXzdiHjuhqDlz1vJuLJ60P7TY,2559
+chatterer/utils/cli.py,sha256=F-Ooz4_pFeX4Y36R7OGhNrE7-IDyVIv2EEelYwJ4H0Y,19392
+chatterer/utils/code_agent.py,sha256=3Wu0M6VvfvWKu4UUoJ0rV-WjvLj06r15_01vft3d1Ns,10231
+chatterer/utils/imghdr.py,sha256=aZ1_AsRzyTsbV7uoeAZMVaC-hj73kvnFHMdHtFKskdE,3694
+chatterer-0.1.14.dist-info/METADATA,sha256=nMitUtZcmWPpmwVl_MTuzvK_AK-t6CuFlsdIxiSmbfA,11267
+chatterer-0.1.14.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
+chatterer-0.1.14.dist-info/top_level.txt,sha256=7nSQKP0bHxPRc7HyzdbKsJdkvPgYD0214o6slRizv9s,10
+chatterer-0.1.14.dist-info/RECORD,,

chatterer/tools/webpage_to_markdown/__init__.py

@@ -1,4 +0,0 @@
-from .playwright_bot import PlayWrightBot
-from .utils import MarkdownLink
-
-__all__ = ["PlayWrightBot", "MarkdownLink"]

chatterer-0.1.12.dist-info/METADATA

@@ -1,170 +0,0 @@
-Metadata-Version: 2.4
-Name: chatterer
-Version: 0.1.12
-Summary: The highest-level interface for various LLM APIs.
-Requires-Python: >=3.12
-Description-Content-Type: text/markdown
-Requires-Dist: instructor>=1.7.2
-Requires-Dist: langchain>=0.3.19
-Provides-Extra: dev
-Requires-Dist: neo4j-extension>=0.1.14; extra == "dev"
-Requires-Dist: colorama>=0.4.6; extra == "dev"
-Requires-Dist: ipykernel>=6.29.5; extra == "dev"
-Provides-Extra: conversion
-Requires-Dist: markdownify>=1.1.0; extra == "conversion"
-Requires-Dist: commonmark>=0.9.1; extra == "conversion"
-Requires-Dist: playwright>=1.50.0; extra == "conversion"
-Requires-Dist: pillow>=11.1.0; extra == "conversion"
-Requires-Dist: mistune>=3.1.2; extra == "conversion"
-Requires-Dist: markitdown>=0.0.2; extra == "conversion"
-Requires-Dist: pymupdf>=1.25.4; extra == "conversion"
-Requires-Dist: youtube-transcript-api>=1.0.2; extra == "conversion"
-Provides-Extra: langchain
-Requires-Dist: chatterer[langchain-providers]; extra == "langchain"
-Requires-Dist: langchain-experimental>=0.3.4; extra == "langchain"
-Provides-Extra: langchain-providers
-Requires-Dist: langchain-openai>=0.3.7; extra == "langchain-providers"
-Requires-Dist: langchain-anthropic>=0.3.8; extra == "langchain-providers"
-Requires-Dist: langchain-google-genai>=2.0.10; extra == "langchain-providers"
-Requires-Dist: langchain-ollama>=0.2.3; extra == "langchain-providers"
-Provides-Extra: all
-Requires-Dist: chatterer[langchain]; extra == "all"
-Requires-Dist: chatterer[conversion]; extra == "all"
-Requires-Dist: chatterer[dev]; extra == "all"
-
-# Chatterer
-
-**Simplified, Structured AI Assistant Framework**
-
-`chatterer` is a Python library designed as a type-safe LangChain wrapper for interacting with various language models (OpenAI, Anthropic, Gemini, Ollama, etc.). It supports structured outputs via Pydantic models, plain text responses, and asynchronous calls.
-
-The structured reasoning in `chatterer` is inspired by the [Atom-of-Thought](https://github.com/qixucen/atom) pipeline.
-
----
-
-## Quick Install
-
-```bash
-pip install chatterer
-```
-
----
-
-## Quickstart Example
-
-Generate text quickly using OpenAI:
-
-```python
-from chatterer import Chatterer
-
-chat = Chatterer.openai("gpt-4o-mini")
-response = chat.generate("What is the meaning of life?")
-print(response)
-```
-
-Messages can be input as plain strings or structured lists:
-
-```python
-response = chat.generate([{ "role": "user", "content": "What's 2+2?" }])
-print(response)
-```
-
-### Structured Output with Pydantic
-
-```python
-from pydantic import BaseModel
-
-class AnswerModel(BaseModel):
-    question: str
-    answer: str
-
-response = chat.generate_pydantic(AnswerModel, "What's the capital of France?")
-print(response.question, response.answer)
-```
-
-### Async Example
-
-```python
-import asyncio
-
-async def main():
-    response = await chat.agenerate("Explain async in Python briefly.")
-    print(response)
-
-asyncio.run(main())
-```
-
----
-
-## Atom-of-Thought Pipeline (AoT)
-
-`AoTPipeline` provides structured reasoning by:
-
-- Detecting question domains (general, math, coding, philosophy, multihop).
-- Decomposing questions recursively.
-- Generating direct, decomposition-based, and simplified answers.
-- Combining answers via ensemble.
-
-### AoT Usage Example
-
-```python
-from chatterer import Chatterer
-from chatterer.strategies import AoTStrategy, AoTPipeline
-
-pipeline = AoTPipeline(chatterer=Chatterer.openai(), max_depth=2)
-strategy = AoTStrategy(pipeline=pipeline)
-
-question = "What would Newton discover if hit by an apple falling from 100 meters?"
-answer = strategy.invoke(question)
-print(answer)
-```
-
----
-
-## Supported Models
-
-- **OpenAI**
-- **Anthropic**
-- **Google Gemini**
-- **Ollama** (local models)
-
-Initialize models easily:
-
-```python
-openai_chat = Chatterer.openai("gpt-4o-mini")
-anthropic_chat = Chatterer.anthropic("claude-3-7-sonnet-20250219")
-gemini_chat = Chatterer.google("gemini-2.0-flash")
-ollama_chat = Chatterer.ollama("deepseek-r1:1.5b")
-```
-
----
-
-## Advanced Features
-
-- **Streaming responses**
-- **Async/Await support**
-- **Structured outputs with Pydantic models**
-
----
-
-## Logging
-
-Built-in logging for easy debugging:
-
-```python
-import logging
-logging.basicConfig(level=logging.DEBUG)
-```
-
----
-
-## Contributing
-
-Feel free to open an issue or pull request.
-
----
-
-## License
-
-MIT License
-

chatterer-0.1.12.dist-info/RECORD

@@ -1,27 +0,0 @@
-chatterer/__init__.py,sha256=dgmZGLN_oJDi6zgxxtNMJz5KKA_bSZ11l8cP9f54XhA,1328
-chatterer/language_model.py,sha256=mT_PNvXOKSvf9SezJO3EQKYJnfP2XwRObIdFSn99L2Q,23501
-chatterer/messages.py,sha256=rk2rto4eDdkKoXe9w8C0TbYoNb4FuEVwGc_uiLPDLOQ,220
-chatterer/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-chatterer/strategies/__init__.py,sha256=oroDpp5ppWralCER5wnf8fgX77dqLAPF0ogmRtRzQfU,208
-chatterer/strategies/atom_of_thoughts.py,sha256=g801rY7k5UeNOq2-XRqB4h8sXjikVNzN57G6OtiUH00,39897
-chatterer/strategies/base.py,sha256=rqOzTo6S2eQ3A_F9aBXCmVoLM1eT6F2VzZ2Dof330Tk,413
-chatterer/tools/__init__.py,sha256=JgASkLhhy0H2coF5toZGDphI5vGtxfSIrLHYGSK2Hdk,677
-chatterer/tools/convert_to_text.py,sha256=isRJ667sEXtZWqpgRx2koIaC3dIP6Pe1_PJXuR10TtQ,15452
-chatterer/tools/youtube.py,sha256=P6I4deJt4xRuQ2ASLGELfkedD1c-foxjIM0n8RI9YTk,5891
-chatterer/tools/citation_chunking/__init__.py,sha256=DyLMGG4dVgSnGIdaSHcBNDz09iXflcKuJCtg4W0JTVo,79
-chatterer/tools/citation_chunking/chunks.py,sha256=_Sxzfbud8XTOHdHdQmKwm4-byES1cD1V6C28DgtS1BA,2120
-chatterer/tools/citation_chunking/citation_chunker.py,sha256=Aye1BqUCa4u_CsTZoqCe72pJA8C_y2U5UR7cNoNpeo4,4776
-chatterer/tools/citation_chunking/citations.py,sha256=BWhSwzZccvu0Db-OxEbsuEGEz-Dh_tXo8HRx1y2XUHg,12308
-chatterer/tools/citation_chunking/prompt.py,sha256=so-8uFQ5b2Zq2V5Brfxd76bEnKYkHovYsohAnbxWEnY,7557
-chatterer/tools/citation_chunking/reference.py,sha256=m47XYaB5uFff_x_k7US9hNr-SpZjKnl-GuzsGaQzcZo,893
-chatterer/tools/citation_chunking/utils.py,sha256=Xytm9lMrS783Po1qWAdEJ8q7Q3l2UMzwHd9EkYTRiwk,6210
-chatterer/tools/webpage_to_markdown/__init__.py,sha256=aL4O78CX6AVBXVfUoM8gLxfYb-kpmhwzwDxKk1Gj_Co,119
-chatterer/tools/webpage_to_markdown/playwright_bot.py,sha256=U5xvGFzn4IWCAI98hpbCD5imKRrhYBa9_RzuFNiSwYA,26847
-chatterer/tools/webpage_to_markdown/utils.py,sha256=N-eZrLQMEmrBwlBBNB1t7c6eEWLGi2kdqoYyrqYq-ec,12604
-chatterer/utils/__init__.py,sha256=qf1gGonmQP3EAaMJtZf_BCYtxgrKH8ymdvD9sF_ezes,319
-chatterer/utils/code_agent.py,sha256=jfv7a67hwHpdbGg8MmjHrZ3buRE2r32XVlXaWtqtuKY,5005
-chatterer/utils/image.py,sha256=nRnHXYtU9LVm6lChvvFl67GytkmD4t6xahbCJprThvQ,10661
-chatterer-0.1.12.dist-info/METADATA,sha256=61l16uSGfXUfUCcXfGnYvYSlVcvrPkRyedhAoiavyVo,4288
-chatterer-0.1.12.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
-chatterer-0.1.12.dist-info/top_level.txt,sha256=7nSQKP0bHxPRc7HyzdbKsJdkvPgYD0214o6slRizv9s,10
-chatterer-0.1.12.dist-info/RECORD,,