Archives
/
langchain
mirror of https://github.com/hwchase17/langchain
2
0
Fork 1
Code Issues Packages Projects Releases Wiki Activity

docs: add agent prompt creation examples (#15957)

Browse Source
pull/15964/head
Bagatur 5 months ago committed by GitHub
parent 69533c8628
commit c697c89ca4
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
8 changed files with 353 additions and 83 deletions
Show Stats Download Patch File Download Diff File

16
docs/docs/modules/agents/agent_types/index.mdx
Unescape Escape View File

@ -30,12 +30,12 @@ Whether this agent requires the model to support any additional parameters. Some
Our commentary on when you should consider using this agent type.
| Agent Type | Intended Model Type | Supports Chat History | Supports Multi-Input Tools | Supports Parallel Function Calling | Required Model Params | When to Use |
| Agent Type | Intended Model Type | Supports Chat History | Supports Multi-Input Tools | Supports Parallel Function Calling | Required Model Params | When to Use | API |
|--------------------------------------------|---------------------|-----------------------|----------------------------|-------------------------------------|----------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------|
| [OpenAI Tools](./openai_tools) | Chat | ✅ | ✅ | ✅ | `tools` | If you are using a recent OpenAI model (`1106` onwards) |
| [OpenAI Functions](./openai_functions_agent)| Chat | ✅ | ✅ | | `functions` | If you are using an OpenAI model, or an open-source model that has been finetuned for function calling and exposes the same `functions` parameters as OpenAI |
| [XML](./xml_agent) | LLM | ✅ | | | | If you are using Anthropic models, or other models good at XML |
| [Structured Chat](./structured_chat) | Chat | ✅ | ✅ | | | If you need to support tools with multiple inputs |
| [JSON Chat](./json_agent) | Chat | ✅ | | | | If you are using a model good at JSON |
| [ReAct](./react) | LLM | ✅ | | | | If you are using a simple model |
| [Self Ask With Search](./self_ask_with_search)| LLM | | | | | If you are using a simple model and only have one search tool |
| [OpenAI Tools](./openai_tools) | Chat | ✅ | ✅ | ✅ | `tools` | If you are using a recent OpenAI model (`1106` onwards) | [Ref](https://api.python.langchain.com/en/latest/agents/langchain.agents.openai_tools.base.create_openai_tools_agent.html) |
| [OpenAI Functions](./openai_functions_agent)| Chat | ✅ | ✅ | | `functions` | If you are using an OpenAI model, or an open-source model that has been finetuned for function calling and exposes the same `functions` parameters as OpenAI | [Ref](https://api.python.langchain.com/en/latest/agents/langchain.agents.openai_functions_agent.base.create_openai_functions_agent.html) |
| [XML](./xml_agent) | LLM | ✅ | | | | If you are using Anthropic models, or other models good at XML | [Ref](https://api.python.langchain.com/en/latest/agents/langchain.agents.xml.base.create_xml_agent.html) |
| [Structured Chat](./structured_chat) | Chat | ✅ | ✅ | | | If you need to support tools with multiple inputs | [Ref](https://api.python.langchain.com/en/latest/agents/langchain.agents.structured_chat.base.create_structured_chat_agent.html) |
| [JSON Chat](./json_agent) | Chat | ✅ | | | | If you are using a model good at JSON | [Ref](https://api.python.langchain.com/en/latest/agents/langchain.agents.json_chat.base.create_json_chat_agent.html) |
| [ReAct](./react) | LLM | ✅ | | | | If you are using a simple model | [Ref](https://api.python.langchain.com/en/latest/agents/langchain.agents.react.agent.create_react_agent.html) |
| [Self Ask With Search](./self_ask_with_search)| LLM | | | | | If you are using a simple model and only have one search tool | [Ref](https://api.python.langchain.com/en/latest/agents/langchain.agents.self_ask_with_search.base.create_self_ask_with_search_agent.html) |

98
libs/langchain/langchain/agents/json_chat/base.py
Unescape Escape View File

@ -16,8 +16,20 @@ def create_json_chat_agent(
) -> Runnable:
"""Create an agent that uses JSON to format its logic, build for Chat Models.
Examples:
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input keys:
`tools`: contains descriptions and arguments for each tool.
`tool_names`: contains all tool names.
`agent_scratchpad`: contains previous agent actions and tool outputs.
Returns:
A Runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
Example:
.. code-block:: python
@ -46,18 +58,82 @@ def create_json_chat_agent(
}
)
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input keys of
`tools`, `tool_names`, and `agent_scratchpad`.
Creating prompt example:
Returns:
A runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
.. code-block:: python
"""
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
system = '''Assistant is a large language model trained by OpenAI.
Assistant is designed to be able to assist with a wide range of tasks, from answering \
simple questions to providing in-depth explanations and discussions on a wide range of \
topics. As a language model, Assistant is able to generate human-like text based on \
the input it receives, allowing it to engage in natural-sounding conversations and \
provide responses that are coherent and relevant to the topic at hand.
Assistant is constantly learning and improving, and its capabilities are constantly \
evolving. It is able to process and understand large amounts of text, and can use this \
knowledge to provide accurate and informative responses to a wide range of questions. \
Additionally, Assistant is able to generate its own text based on the input it \
receives, allowing it to engage in discussions and provide explanations and \
descriptions on a wide range of topics.
Overall, Assistant is a powerful system that can help with a wide range of tasks \
and provide valuable insights and information on a wide range of topics. Whether \
you need help with a specific question or just want to have a conversation about \
a particular topic, Assistant is here to assist.'''
human = '''TOOLS
------
Assistant can ask the user to use tools to look up information that may be helpful in \
answering the users original question. The tools the human can use are:
{tools}
RESPONSE FORMAT INSTRUCTIONS
----------------------------
When responding to me, please output a response in one of two formats:
**Option 1:**
Use this if you want the human to use a tool.
Markdown code snippet formatted in the following schema:
```json
{{
"action": string, \ The action to take. Must be one of {tool_names}
"action_input": string \ The input to the action
}}
```
**Option #2:**
Use this if you want to respond directly to the human. Markdown code snippet formatted \
in the following schema:
```json
{{
"action": "Final Answer",
"action_input": string \ You should put what you want to return to use here
}}
```
USER'S INPUT
--------------------
Here is the user's input (remember to respond with a markdown code snippet of a json \
blob with a single action, and NOTHING else):
{input}'''
prompt = ChatPromptTemplate.from_messages(
[
("system", system),
MessagesPlaceholder("chat_history", optional=True),
("human", human),
MessagesPlaceholder("agent_scratchpad"),
]
)
""" # noqa: E501
missing_vars = {"tools", "tool_names", "agent_scratchpad"}.difference(
prompt.input_variables
)

37
libs/langchain/langchain/agents/openai_functions_agent/base.py
Unescape Escape View File

@ -235,7 +235,20 @@ def create_openai_functions_agent(
) -> Runnable:
"""Create an agent that uses OpenAI function calling.
Examples:
Args:
llm: LLM to use as the agent. Should work with OpenAI function calling,
so either be an OpenAI model that supports that or a wrapper of
a different model that adds in equivalent support.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input key `agent_scratchpad`, which will
contain agent action and tool output messages.
Returns:
A Runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
Example:
Creating an agent with no memory
@ -266,18 +279,20 @@ def create_openai_functions_agent(
}
)
Args:
llm: LLM to use as the agent. Should work with OpenAI function calling,
so either be an OpenAI model that supports that or a wrapper of
a different model that adds in equivalent support.
tools: Tools this agent has access to.
prompt: The prompt to use, must have an input key of `agent_scratchpad`.
Creating prompt example:
Returns:
A runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
.. code-block:: python
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
prompt = ChatPromptTemplate.from_messages(
[
("system", "You are a helpful assistant"),
MessagesPlaceholder("chat_history", optional=True),
("human", "{input}"),
MessagesPlaceholder("agent_scratchpad"),
]
)
"""
if "agent_scratchpad" not in prompt.input_variables:
raise ValueError(

34
libs/langchain/langchain/agents/openai_tools/base.py
Unescape Escape View File

@ -17,7 +17,18 @@ def create_openai_tools_agent(
) -> Runnable:
"""Create an agent that uses OpenAI tools.
Examples:
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input key `agent_scratchpad`, which will
contain agent action and tool output messages.
Returns:
A Runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
Example:
.. code-block:: python
@ -46,15 +57,20 @@ def create_openai_tools_agent(
}
)
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input keys of `agent_scratchpad`.
Creating prompt example:
Returns:
A runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
.. code-block:: python
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
prompt = ChatPromptTemplate.from_messages(
[
("system", "You are a helpful assistant"),
MessagesPlaceholder("chat_history", optional=True),
("human", "{input}"),
MessagesPlaceholder("agent_scratchpad"),
]
)
"""
missing_vars = {"agent_scratchpad"}.difference(prompt.input_variables)
if missing_vars:

50
libs/langchain/langchain/agents/react/agent.py
Unescape Escape View File

@ -17,6 +17,20 @@ def create_react_agent(
) -> Runnable:
"""Create an agent that uses ReAct prompting.
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input keys:
`tools`: contains descriptions and arguments for each tool.
`tool_names`: contains all tool names.
`agent_scratchpad`: contains previous agent actions and tool outputs.
Returns:
A Runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
Examples:
.. code-block:: python
@ -45,18 +59,34 @@ def create_react_agent(
}
)
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input keys of
`tools`, `tool_names`, and `agent_scratchpad`.
Creating prompt example:
Returns:
A runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
.. code-block:: python
from langchain_core.prompts import PromptTemplate
template = '''Answer the following questions as best you can. You have access to the following tools:
{tools}
Use the following format:
Question: the input question you must answer
Thought: you should always think about what to do
Action: the action to take, should be one of [{tool_names}]
Action Input: the input to the action
Observation: the result of the action
... (this Thought/Action/Action Input/Observation can repeat N times)
Thought: I now know the final answer
Final Answer: the final answer to the original input question
Begin!
Question: {input}
Thought:{agent_scratchpad}'''
"""
prompt = PromptTemplate.from_template(template)
""" # noqa: E501
missing_vars = {"tools", "tool_names", "agent_scratchpad"}.difference(
prompt.input_variables
)

68
libs/langchain/langchain/agents/self_ask_with_search/base.py
Unescape Escape View File

@ -91,8 +91,19 @@ def create_self_ask_with_search_agent(
) -> Runnable:
"""Create an agent that uses self-ask with search prompting.
Examples:
Args:
llm: LLM to use as the agent.
tools: List of tools. Should just be of length 1, with that tool having
name `Intermediate Answer`
prompt: The prompt to use, must have input key `agent_scratchpad` which will
contain agent actions and tool outputs.
Returns:
A Runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
Examples:
.. code-block:: python
@ -111,18 +122,53 @@ def create_self_ask_with_search_agent(
agent_executor.invoke({"input": "hi"})
Args:
llm: LLM to use as the agent.
tools: List of tools. Should just be of length 1, with that tool having
name `Intermediate Answer`
prompt: The prompt to use, must have input keys of `agent_scratchpad`.
Create prompt example:
Returns:
A runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
.. code-block:: python
"""
from langchain_core.prompts import PromptTemplate
template = '''Question: Who lived longer, Muhammad Ali or Alan Turing?
Are follow up questions needed here: Yes.
Follow up: How old was Muhammad Ali when he died?
Intermediate answer: Muhammad Ali was 74 years old when he died.
Follow up: How old was Alan Turing when he died?
Intermediate answer: Alan Turing was 41 years old when he died.
So the final answer is: Muhammad Ali
Question: When was the founder of craigslist born?
Are follow up questions needed here: Yes.
Follow up: Who was the founder of craigslist?
Intermediate answer: Craigslist was founded by Craig Newmark.
Follow up: When was Craig Newmark born?
Intermediate answer: Craig Newmark was born on December 6, 1952.
So the final answer is: December 6, 1952
Question: Who was the maternal grandfather of George Washington?
Are follow up questions needed here: Yes.
Follow up: Who was the mother of George Washington?
Intermediate answer: The mother of George Washington was Mary Ball Washington.
Follow up: Who was the father of Mary Ball Washington?
Intermediate answer: The father of Mary Ball Washington was Joseph Ball.
So the final answer is: Joseph Ball
Question: Are both the directors of Jaws and Casino Royale from the same country?
Are follow up questions needed here: Yes.
Follow up: Who is the director of Jaws?
Intermediate answer: The director of Jaws is Steven Spielberg.
Follow up: Where is Steven Spielberg from?
Intermediate answer: The United States.
Follow up: Who is the director of Casino Royale?
Intermediate answer: The director of Casino Royale is Martin Campbell.
Follow up: Where is Martin Campbell from?
Intermediate answer: New Zealand.
So the final answer is: No
Question: {input}
Are followup questions needed here:{agent_scratchpad}'''
prompt = PromptTemplate.from_template(template)
""" # noqa: E501
missing_vars = {"agent_scratchpad"}.difference(prompt.input_variables)
if missing_vars:
raise ValueError(f"Prompt missing required variables: {missing_vars}")

79
libs/langchain/langchain/agents/structured_chat/base.py
Unescape Escape View File

@ -155,8 +155,20 @@ def create_structured_chat_agent(
) -> Runnable:
"""Create an agent aimed at supporting tools with multiple inputs.
Examples:
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input keys
`tools`: contains descriptions and arguments for each tool.
`tool_names`: contains all tool names.
`agent_scratchpad`: contains previous agent actions and tool outputs.
Returns:
A Runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
Examples:
.. code-block:: python
@ -185,18 +197,63 @@ def create_structured_chat_agent(
}
)
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input keys of
`tools`, `tool_names`, and `agent_scratchpad`.
Creating prompt example:
Returns:
A runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
.. code-block:: python
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
system = '''Respond to the human as helpfully and accurately as possible. You have access to the following tools:
{tools}
Use a json blob to specify a tool by providing an action key (tool name) and an action_input key (tool input).
Valid "action" values: "Final Answer" or {tool_names}
Provide only ONE action per $JSON_BLOB, as shown:
```
{{
"action": $TOOL_NAME,
"action_input": $INPUT
}}
```
"""
Follow this format:
Question: input question to answer
Thought: consider previous and subsequent steps
Action:
```
$JSON_BLOB
```
Observation: action result
... (repeat Thought/Action/Observation N times)
Thought: I know what to respond
Action:
```
{{
"action": "Final Answer",
"action_input": "Final response to human"
}}
Begin! Reminder to ALWAYS respond with a valid json blob of a single action. Use tools if necessary. Respond directly if appropriate. Format is Action:```$JSON_BLOB```then Observation'''
human = '''{input}
{agent_scratchpad}
(reminder to respond in a JSON blob no matter what)'''
prompt = ChatPromptTemplate.from_messages(
[
("system", system),
MessagesPlaceholder("chat_history", optional=True),
("human", human),
]
)
""" # noqa: E501
missing_vars = {"tools", "tool_names", "agent_scratchpad"}.difference(
prompt.input_variables
)

54
libs/langchain/langchain/agents/xml/base.py
Unescape Escape View File

@ -112,8 +112,19 @@ def create_xml_agent(
) -> Runnable:
"""Create an agent that uses XML to format its logic.
Examples:
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input keys
`tools`: contains descriptions for each tool.
`agent_scratchpad`: contains previous agent actions and tool outputs.
Returns:
A Runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
Example:
.. code-block:: python
@ -137,22 +148,41 @@ def create_xml_agent(
"input": "what's my name?",
# Notice that chat_history is a string
# since this prompt is aimed at LLMs, not chat models
"chat_history": "Human: My name is Bob\nAI: Hello Bob!",
"chat_history": "Human: My name is Bob\\nAI: Hello Bob!",
}
)
Args:
llm: LLM to use as the agent.
tools: Tools this agent has access to.
prompt: The prompt to use, must have input keys of
`tools` and `agent_scratchpad`.
Creating prompt example:
Returns:
A runnable sequence representing an agent. It takes as input all the same input
variables as the prompt passed in does. It returns as output either an
AgentAction or AgentFinish.
.. code-block:: python
"""
from langchain_core.prompts import PromptTemplate
template = '''You are a helpful assistant. Help the user answer any questions.
You have access to the following tools:
{tools}
In order to use a tool, you can use <tool></tool> and <tool_input></tool_input> tags. You will then get back a response in the form <observation></observation>
For example, if you have a tool called 'search' that could run a google search, in order to search for the weather in SF you would respond:
<tool>search</tool><tool_input>weather in SF</tool_input>
<observation>64 degrees</observation>
When you are done, respond with a final answer between <final_answer></final_answer>. For example:
<final_answer>The weather in SF is 64 degrees</final_answer>
Begin!
Previous Conversation:
{chat_history}
Question: {input}
{agent_scratchpad}'''
prompt = PromptTemplate.from_template(template)
""" # noqa: E501
missing_vars = {"tools", "agent_scratchpad"}.difference(prompt.input_variables)
if missing_vars:
raise ValueError(f"Prompt missing required variables: {missing_vars}")

Write Preview
Loading…
Cancel
Save