Merge pull request #4 from hwchase17/readme-update

update readmes

README.md

@@ -2,53 +2,31 @@
 
 [Warning: very beta, may change drastically]
 
-Taking inspiration from Hugging Face Hub, this is collection of all artifacts useful for working with LangChain chains.
-To start, this focuses on prompts.
-This is intended to be a central place to upload and share prompts.
-It is intended to be very community driven and we hope that people will contribute their own prompts.
-Please see below for instructions on loading and uploading prompts.
+Taking inspiration from Hugging Face Hub, LangChainHub is collection of all artifacts useful for working with LangChain primitives such as prompts, chains and agents. 
+The goal of this repository is to be a central resource for sharing and discovering high quality prompts, chains and agents that combine together to form complex LLM applications.
 
-## Loading
+We are starting off the hub with a collection of prompts, and we look forward to the LangChain community adding to this collection. We hope to expand to chains and agents shortly.
 
-All prompts can be loaded from LangChain by specifying the desired path.
-The path should be relative to the `prompts` folder here.
-For example, if there is a file at `prompts/qa/stuff/basic/prompt.yaml`, the path you want to specify is `qa/stuff/basic/prompt.yaml`
+## 📖 Prompts
 
-Once you have that path, you can load it in the following manner:
+At a high level, prompts are organized by use case inside the `prompts` directory.
+To load a prompt in LangChain, you should use the following code snippet:
 
 ```python
-from langchain.prompts import load_from_hub
-prompt = load_from_hub("qa/stuff/basic/prompt.yaml")
-```
+from langchain.prompts import load_prompt
 
-## Uploading
+prompt = load_prompt('lc://prompts/path/to/file.json')
+```
 
-There are three methods for uploading prompts: `json`, `yaml`, and `python`.
-The suggested options are `json` and `yaml`, but we provide `python` as an option for more flexibility.
-Please see the below sections for instructions for uploading each format.
+In addition to prompt files themselves, each sub-directory also contains a README explaining how best to use that prompt in the appropriate LangChain chain.
 
-You should upload your prompt file to a folder in the appropriate use case section.
-In addition to the prompt file, you should also add a README.md for that folder, including any relevant information about prompt.
-This can be how it was created, what it should be used for, what the variables mean, etc.
+For more detailed information on how prompts are organized in the Hub, and how best to upload one, please see the documentation [here](./prompts/README.md)
 
-### `json`
-To get a properly formatted json file, if you have prompt in memory in Python you can run:
-```python
-prompt.save("file_name.json")
-```
+## 🔗 Chains
 
-Replace `"file_name"` with the desired name of the file.
-
-### `yaml`
-To get a properly formatted yaml file, if you have prompt in memory in Python you can run:
-```python
-prompt.save("file_name.yaml")
-```
+Coming soon!
 
-Replace `"file_name"` with the desired name of the file.
 
+## 🤖 Agents
 
-### `python`
-To get a properly formatted Python file, you should upload a Python file that exposes a `PROMPT` variable.
-This is the variable that will be loaded.
-This variable should be an instance of a subclass of BasePromptTemplate in LangChain.
+Coming soon!

prompts/README.md

@@ -0,0 +1,60 @@
+# Prompts
+
+This directory covers loading and uploading of prompts. 
+Each sub-directory covers a different use case, and has not only relevant prompts for that use case but also a README file describing how to best use that prompt.
+
+## Loading
+
+All prompts can be loaded from LangChain by specifying the desired path, and adding the `lc://` prefix. The path should be relative to the `langchain-hub` repo.
+
+For example, to load the prompt at the path `langchain-hub/prompts/qa/stuff/basic/prompt.yaml`, the path you want to specify is `lc://prompts/qa/stuff/basic/prompt.yaml`
+
+Once you have that path, you can load it in the following manner:
+
+```python
+from langchain.prompts import load_prompt
+
+prompt = load_prompt('lc://prompts/qa/stuff/basic/prompt.yaml')
+```
+
+## Uploading
+
+To upload a prompt to the LangChainHub, you must upload 2 files:
+1. The prompt. There are 3 supported file formats for prompts: `json`, `yaml`, and `python`. The suggested options are `json` and `yaml`, but we provide `python` as an option for more flexibility. Please see the below sections for instructions for uploading each format.
+2. Associated README file for the prompt. This provides a high level description of the prompt, usage patterns of the prompt and chains that the prompt is compatible with. For more details, check out langchain-hub/readme_template.
+   If you are uploading a prompt to an existing directory, it should already have a README file and so this should not be necessary.
+
+
+The prompts on the hub are organized by use case. The use cases are reflected in the directory structure and names, and each separate directory represents a different use case. You should upload your prompt file to a folder in the appropriate use case section.
+
+
+If adding a prompt to an existing use case folder, then make sure that the prompt:
+1. services the same use case as the existing prompt(s) in that folder, and
+2. has the same inputs as the existing prompt(s).
+
+A litmus test to make sure that multiple prompts belong in the same folder: the existing README file for that folder should also apply to the new prompt being added.
+
+
+### Supported file formats
+
+#### `json`
+To get a properly formatted json file, if you have prompt in memory in Python you can run:
+```python
+prompt.save("file_name.json")
+```
+
+Replace `"file_name"` with the desired name of the file.
+
+#### `yaml`
+To get a properly formatted yaml file, if you have prompt in memory in Python you can run:
+```python
+prompt.save("file_name.yaml")
+```
+
+Replace `"file_name"` with the desired name of the file.
+
+
+#### `python`
+To get a properly formatted Python file, you should upload a Python file that exposes a `PROMPT` variable.
+This is the variable that will be loaded.
+This variable should be an instance of a subclass of BasePromptTemplate in LangChain.

prompts/api/api_response/README.md

@@ -17,12 +17,12 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import APIChain
 
 llm = ...
 api_docs = ...
-prompt = load_from_hub('api/api_response/<file-name>')
+prompt = load_prompt('lc://prompts/api/api_response/<file-name>')
 chain = APIChain.from_llm_and_api_docs(llm, api_docs, api_response_prompt=prompt)
 ```
 

prompts/api/api_url/README.md

@@ -15,12 +15,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import APIChain
 
 llm = ...
 api_docs = ...
-prompt = load_from_hub('api/api_url/<file-name>')
+prompt = load_prompt('lc://prompts/api/api_url/<file-name>')
 chain = APIChain.from_llm_and_api_docs(llm, api_docs, api_url_prompt=prompt)
 ```
-

prompts/conversation/README.md

@@ -16,11 +16,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import ConversationChain
 
 llm = ...
-prompt = load_from_hub('conversation/<file-name>')
+prompt = load_prompt('lc://prompts/conversation/<file-name>')
 chain = ConversationChain(llm=llm, prompt=prompt)
 ```
 

prompts/hello-world/README.md

@@ -3,16 +3,21 @@
 Basic prompt designed to be use as a test case, will just instruct the LLM to say "Hello World".
 
 
+## Inputs
+
+This prompt doesn't have any inputs.
+
+
 ## Usage
 
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import LLMChain
 
 llm = ...
-prompt = load_from_hub('hello-world/<file-name>')
+prompt = load_prompt('lc://prompts/hello-world/<file-name>')
 chain = LLMChain(llm=llm, prompt=prompt)
 ```
 

prompts/llm_bash/README.md

@@ -14,11 +14,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import LLMBashChain
 
 llm = ...
-prompt = load_from_hub('llm_bash/<file-name>')
+prompt = load_prompt('lc://prompts/llm_bash/<file-name>')
 chain = LLMBashChain(llm=llm, prompt=prompt)
 ```
 

prompts/llm_math/README.md

@@ -15,11 +15,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import LLMMathChain
 
 llm = ...
-prompt = load_from_hub('llm_math/<file-name>')
+prompt = load_prompt('lc://prompts/llm_math/<file-name>')
 chain = LLMMathChain(llm=llm, prompt=prompt)
 ```
 

prompts/memory/summarize/README.md

@@ -16,12 +16,12 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import ConversationChain
 from langchain.chains.conversation.memory import ConversationSummaryMemory
 
 llm = ...
-prompt = load_from_hub('memory/summarize/<file-name>')
+prompt = load_prompt('lc://prompts/memory/summarize/<file-name>')
 memory = ConversationSummaryMemory(llm=llm, prompt=prompt)
 chain = ConversationChain(llm=llm, memory=memory)
 ```

prompts/pal/README.md

@@ -16,13 +16,13 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import PALChain
 
 llm = ...
 stop = ...
 get_answer_expr = ...
-prompt = load_from_hub('pal/<file-name>')
+prompt = load_prompt('lc://prompts/pal/<file-name>')
 chain = PALChain(llm=llm, prompt=prompt, stop=stop, get_answer_expr=get_answer_expr)
 ```
 

prompts/qa/map_reduce/question/README.md

@@ -16,11 +16,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.question_answering import load_qa_chain
 
 llm = ...
-prompt = load_from_hub('qa/map_reduce/question/<file-name>')
+prompt = load_prompt('lc://prompts/qa/map_reduce/question/<file-name>')
 chain = load_qa_chain(llm, chain_type="map_reduce", question_prompt=prompt)
 ```
 

prompts/qa/map_reduce/reduce/README.md

@@ -16,11 +16,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.question_answering import load_qa_chain
 
 llm = ...
-prompt = load_from_hub('qa/map_reduce/reduce/<file-name>')
+prompt = load_prompt('lc://prompts/qa/map_reduce/reduce/<file-name>')
 chain = load_qa_chain(llm, chain_type="map_reduce", combine_prompt=prompt)
 ```
 

prompts/qa/refine/README.md

@@ -17,11 +17,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.question_answering import load_qa_chain
 
 llm = ...
-prompt = load_from_hub('qa/refine/<file-name>')
+prompt = load_prompt('lc://prompts/qa/refine/<file-name>')
 chain = load_qa_chain(llm, chain_type="refine", refine_prompt=prompt)
 ```
 

prompts/qa/stuff/README.md

@@ -16,11 +16,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.question_answering import load_qa_chain
 
 llm = ...
-prompt = load_from_hub('qa/stuff/<file-name>')
+prompt = load_prompt('lc://prompts/qa/stuff/<file-name>')
 chain = load_qa_chain(llm, chain_type="stuff", prompt=prompt)
 ```
 

prompts/qa_with_sources/map_reduce/reduce/README.md

@@ -1,20 +1,14 @@
 # Description of QA with Sources Map Reduce Prompts
 
-Prompt
-
-## Compatible Chains
-
-Below is a list of chains we expect this prompt to be compatible with.
-
-1. {{Chain Name}}: {{Path to chain in module}}
-2. ...
+This prompt enables the user to perform question answering while providing sources.
+It uses the map reduce chain for doing QA. This specific prompt reduces the answer generated during the question stage.
 
 ## Inputs
 
 This is a description of the inputs that the prompt expects.
 
-1. {{input_var}}: {{Description}}
-2. ...
+1. `summaries`: Summaries generated during the map step.
+2. `question`: Original question to be answered.
 
 
 ## Usage
@@ -22,11 +16,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.qa_with_sources import load_qa_with_sources_chain
 
 llm = ...
-prompt = load_from_hub('qa_with_sources/map_reduce/reduce/<file-name>')
+prompt = load_prompt('lc://prompts/qa_with_sources/map_reduce/reduce/<file-name>')
 chain = load_qa_with_sources_chain(llm, chain_type="map_reduce", combine_prompt=prompt)
 ```
 

prompts/qa_with_sources/refine/README.md

@@ -16,11 +16,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.qa_with_sources import load_qa_with_sources_chain
 
 llm = ...
-prompt = load_from_hub('qa_with_sources/refine/<file-name>')
+prompt = load_prompt('lc://prompts/qa_with_sources/refine/<file-name>')
 chain = load_qa_with_sources_chain(llm, chain_type="refine", refine_prompt=prompt)
 ```
 

prompts/qa_with_sources/stuff/README.md

@@ -16,11 +16,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.qa_with_sources import load_qa_with_sources_chain
 
 llm = ...
-prompt = load_from_hub('qa_with_sources/stuff/<file-name>')
+prompt = load_prompt('lc://prompts/qa_with_sources/stuff/<file-name>')
 chain = load_qa_with_sources_chain(llm, chain_type="stuff", prompt=prompt)
 ```
 

prompts/sql_query/language_to_sql_output/README.md

@@ -18,12 +18,12 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import SQLDatabaseChain
 
 llm = ...
 database = ...
-prompt = load_from_hub('sql_query/language_to_sql_output/<file-name>')
+prompt = load_prompt('lc://prompts/sql_query/language_to_sql_output/<file-name>')
 chain = SQLDatabaseChain(llm=llm, database=database, prompt=prompt)
 ```
 

prompts/sql_query/relevant_tables/README.md

@@ -16,12 +16,12 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import SQLDatabaseSequentialChain
 
 llm = ...
 database = ...
-prompt = load_from_hub('sql_query/relevant_tables/<file-name>')
+prompt = load_prompt('lc://prompts/sql_query/relevant_tables/<file-name>')
 chain = SQLDatabaseSequentialChain.from_llm(llm, database, decider_prompt=prompt)
 ```
 

prompts/summarize/map_reduce/map/README.md

@@ -15,11 +15,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.summarize import load_summarize_chain
 
 llm = ...
-prompt = load_from_hub('summarize/map_reduce/map/<file-name>')
+prompt = load_prompt('lc://prompts/summarize/map_reduce/map/<file-name>')
 chain = load_summarize_chain(llm, chain_type="map_reduce", map_prompt=prompt)
 ```
 

prompts/summarize/refine/README.md

@@ -16,11 +16,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.summarize import load_summarize_chain
 
 llm = ...
-prompt = load_from_hub('summarize/refine/<file-name>')
+prompt = load_prompt('lc://prompts/summarize/refine/<file-name>')
 chain = load_summarize_chain(llm, chain_type="refine", refine_prompt=prompt)
 ```
 

prompts/summarize/stuff/README.md

@@ -14,11 +14,11 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains.summarize import load_summarize_chain
 
 llm = ...
-prompt = load_from_hub('summarize/stuff/<file-name>')
+prompt = load_prompt('lc://prompts/summarize/stuff/<file-name>')
 chain = load_summarize_chain(llm, chain_type="stuff", prompt=prompt)
 ```
 

prompts/vector_db_qa/README.md

@@ -16,12 +16,12 @@ This is a description of the inputs that the prompt expects.
 Below is a code snippet for how to use the prompt.
 
 ```python
-from langchain.prompts import load_from_hub
+from langchain.prompts import load_prompt
 from langchain.chains import VectorDBQA
 
 llm = ...
 vectorstore = ...
-prompt = load_from_hub('vector_db_qa/<file-name>')
+prompt = load_prompt('lc://prompts/vector_db_qa/<file-name>')
 chain = VectorDBQA.from_llm(llm, prompt=prompt, vectorstore=vectorstore)
 ```