sherpa_ai.actions package

In This Page:

sherpa_ai.actions package#

Submodules#

sherpa_ai.actions.answer_arithmetic module#

sherpa_ai.actions.base module#

class sherpa_ai.actions.base.ActionArgument(*args, name: str, type: str = 'str', description: str = '', source: str = 'agent', key: str | None = None)[source]#

Bases: BaseModel

Argument used for an action.

name#

Name of the argument.

Type:

str

type#

Type of the argument. (default: “str”)

Type:

str

description#

Description of the argument. (default: “”)

Type:

str

source#

source of the argument, select from (agent, belief) If source is agent, the argument is provided by the agent (LLM). If source is belief, the value of the argument is retrieved from the dictionary of the belief. (default: “agent”)

Type:

str

key#

key of the argument in the belief dictionary if source is belief. (default: name of the argument, if source is belief)

Type:

str

description: str#
key: str | None#
model_config: ClassVar[ConfigDict] = {}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str#
source: str#
type: str#
class sherpa_ai.actions.base.ActionResource(*, source: str, content: str)[source]#

Bases: BaseModel

Resource used for an action.

source#

Source of the resource, such as document id or url.

Type:

str

content#

Content of the resource.

Type:

str

content: str#
model_config: ClassVar[ConfigDict] = {}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

source: str#
class sherpa_ai.actions.base.AsyncBaseAction(*args, name: str, usage: str, belief: Any = None, output_key: str | None = None)[source]#

Bases: BaseAction, ABC

async execute(**kwargs)[source]#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class sherpa_ai.actions.base.BaseAction(*args, name: str, usage: str, belief: Any = None, output_key: str | None = None)[source]#

Bases: ABC, BaseModel

Base class for an action.

name#

Name of the action.

Type:

str

args#

Arguments required to run the action.

Type:

Union[dict, list[ActionArgument]]

usage#

Usage description of the action.

Type:

str

belief#

Belief used for the action. It is required if any argument requires belief. (default: None)

Type:

Any

output_key#

Output key of the action. (default: name of the action)

Type:

Optional[str]

class Config[source]#

Bases: object

arbitrary_types_allowed = True#
action_end(result: Any)[source]#
action_start(args: dict)[source]#
args: dict | list[ActionArgument]#
belief: Any#
abstract execute(**kwargs)[source]#
input_validation(**kwargs) → dict[source]#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str#
output_key: str | None#
usage: str#
class sherpa_ai.actions.base.BaseRetrievalAction(*args, name: str, usage: str, belief: ~typing.Any = None, output_key: str | None = None, resources: list[~sherpa_ai.actions.base.ActionResource] = <factory>, num_documents: int = 5, reranker: ~sherpa_ai.actions.utils.reranking.BaseReranking = None, refiner: ~sherpa_ai.actions.utils.refinement.BaseRefinement = None, current_task: str = '', perform_reranking: bool = False, perform_refinement: bool = False)[source]#

Bases: BaseAction, ABC

add_resources(resources: list[dict])[source]#
current_task: str#
execute(query: str) → str[source]#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

num_documents: int#
perform_refinement: bool#
perform_reranking: bool#
refine(documents: list[str]) → list[str][source]#

Refine the results based on the query.

refiner: BaseRefinement#
reranker: BaseReranking#
reranking(documents: list[str]) → list[str][source]#

Rerank the documents based on the query.

resources: list[ActionResource]#
abstract search(query: str) → str[source]#

Search for relevant documents based on the query.

sherpa_ai.actions.deliberation module#

class sherpa_ai.actions.deliberation.Deliberation(*args, name: str = 'Deliberation', usage: str = 'Directly come up with a solution', belief: Any = None, output_key: str | None = None, role_description: str, llm: Any = None, description: str = 'Role Description: {role_description}\nTask Description: {task}\n\nPlease deliberate on the task and generate a solution that is:\n\nHighly Detailed: Break down components and elements clearly.\nQuality-Oriented: Ensure top-notch performance and longevity.\nPrecision-Focused: Specific measures, materials, or methods to be used.\n\nKeep the result concise and short. No more than one paragraph.\n\n')[source]#

Bases: BaseAction

args: dict#
description: str#
execute(task: str) → str[source]#
llm: Any#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str#
role_description: str#
usage: str#

sherpa_ai.actions.planning module#

class sherpa_ai.actions.planning.Plan[source]#

Bases: object

add_step(step: Step)[source]#
classmethod from_dict(data)[source]#
class sherpa_ai.actions.planning.Step(agent_name: str, task: str)[source]#

Bases: object

Step for the plan created by the planner

agent_name: the name of the agent that should execute the step task: the task that the agent should execute

classmethod from_dict(data)[source]#
class sherpa_ai.actions.planning.TaskPlanning(*args, name: str = 'TaskPlanning', usage: str = 'Come up with a plan to solve the task', belief: Any = None, output_key: str | None = None, llm: Any = None, num_steps: int = 5, prompt: str = 'You are a **task decomposition assistant** who simplifies complex tasks into sequential steps, assigning roles or agents to each.\nBy analyzing user-defined tasks and agent capabilities, you provides structured plans, enhancing project clarity and efficiency.\nYour adaptability ensures customized solutions for diverse needs.\n\nA good plan is concise, detailed, feasible and efficient.\n\nTask: **{task}**\n\nAgents:\n{agent_pool_description}\n\nPlease break down the task into maximum {num_steps} individual, detailed steps and designate an appropriate agent for each step. The result should be in the following format:\nStep 1:\n    Agent: <AgentName>\n    Task: <detailed task description>\n...\nStep N:\n    Agent: <AgentName>\n    Task: <detailed task description>\n\nDo not answer anything else, and do not add any other information in your answer. Only select agents from the the list and only select one agent at a time.\n', revision_prompt: str = 'You are a **task decomposition assistant** who simplifies complex tasks into sequential steps, assigning roles or agents to each.\nBy analyzing user-defined tasks and agent capabilities, you provide structured plans, enhancing project clarity and efficiency.\nYour adaptability ensures customized solutions for diverse needs.\n\nA good plan is concise, detailed, feasible and efficient. It should be broken down into individual steps, with each step assigned to an appropriate agent.\n\nTask: **{task}**\n\nAgents:\n{agent_pool_description}\n\nHere is your previous plan:\n{previous_plan}\n\nHere is the feedback from the last run:\n{feedback}\n\nPlease revise the plan based on the feedback to maximum {num_steps} steps. The result should be in the following format:\nStep 1:\n    Agent: <AgentName>\n    Task: <detailed task description>\n...\nStep N:\n    Agent: <AgentName>\n    Task: <detailed task description>\n\nDo not answer anything else, and do not add any other information in your answer. Only select agents from the the list and only select one agent at a time.\n')[source]#

Bases: BaseAction

args: dict#
execute(task: str, agent_pool_description: str, last_plan: str | None = None, feedback: str | None = None) → Plan[source]#

Execute the action

llm: Any#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str#
num_steps: int#
post_process(action_output: str) → Plan[source]#

Post process the action output into a plan with steps

prompt: str#
revision_prompt: str#
usage: str#

sherpa_ai.actions.synthesize module#

class sherpa_ai.actions.synthesize.SynthesizeOutput(*, name: str = 'SynthesizeOutput', args: dict = {'context': 'string', 'history': 'string', 'task': 'string'}, usage: str = 'Answer the question using conversation history with the user', belief: Any = None, output_key: str | None = None, role_description: str, llm: Any = None, description: str = '{role_description}\n\nContext: {context}\n\nAction - Result History:\n{history}\n\nGiven the context and the action-result history, please complete the task mentioned. Include any links you used from the context and history in the result.\nTask: {task}\nResult:\n', add_citation: bool = False)[source]#

Bases: BaseAction

add_citation: bool#
args: dict#
description: str#
execute(task: str, context: str, history: str) → str[source]#
llm: Any#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str#
role_description: str#
usage: str#

Module contents#

class sherpa_ai.actions.ArxivSearch(*, name: str = 'ArxivSearch', args: dict = {'query': 'string'}, usage: str = 'Search paper on the Arxiv website', belief: ~typing.Any = None, output_key: str | None = None, resources: list[~sherpa_ai.actions.base.ActionResource] = <factory>, num_documents: int = 5, reranker: ~sherpa_ai.actions.utils.reranking.BaseReranking = None, refiner: ~sherpa_ai.actions.utils.refinement.BaseRefinement = None, current_task: str = '', perform_reranking: bool = False, perform_refinement: bool = True, role_description: str, task: str, llm: ~typing.Any = None, description: str = 'Role Description: {role_description}\nTask: {task}\n\nRelevant Paper Title and Summary:\n{paper_title_summary}\n\n\nReview and analyze the provided paper summary with respect to the task. Craft a concise and short, unified summary that distills key information that is most relevant to the task, incorporating reference links within the summary.\nOnly use the information given. Do not add any additional information. The summary should be less than {n} setences\n')[source]#

Bases: BaseRetrievalAction

args: dict#
belief: Any#
current_task: str#
description: str#
llm: Any#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_post_init(context: Any, /) → None#

This function is meant to behave like a BaseModel method to initialise private attributes.

It takes context as an argument since that’s what pydantic-core passes when calling it.

Parameters:

  • self – The BaseModel instance.

  • context – The context.

name: str#
num_documents: int#
output_key: str | None#
perform_refinement: bool#
perform_reranking: bool#
refine(result: str) → str[source]#

Refine the results based on the query.

refiner: BaseRefinement#
reranker: BaseReranking#
resources: list[ActionResource]#
role_description: str#
search(query) → list[dict][source]#

Search for relevant documents based on the query.

task: str#
usage: str#
class sherpa_ai.actions.Deliberation(*args, name: str = 'Deliberation', usage: str = 'Directly come up with a solution', belief: Any = None, output_key: str | None = None, role_description: str, llm: Any = None, description: str = 'Role Description: {role_description}\nTask Description: {task}\n\nPlease deliberate on the task and generate a solution that is:\n\nHighly Detailed: Break down components and elements clearly.\nQuality-Oriented: Ensure top-notch performance and longevity.\nPrecision-Focused: Specific measures, materials, or methods to be used.\n\nKeep the result concise and short. No more than one paragraph.\n\n')[source]#

Bases: BaseAction

args: dict#
belief: Any#
description: str#
execute(task: str) → str[source]#
llm: Any#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str#
output_key: str | None#
role_description: str#
usage: str#
class sherpa_ai.actions.EmptyAction(*args, name: str = '', usage: str = 'Make a decision', belief: Any = None, output_key: str | None = None)[source]#

Bases: BaseAction

args: dict#
execute() → str[source]#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str#
usage: str#
class sherpa_ai.actions.GoogleSearch(*, name: str = 'Google Search', args: dict = {'query': 'string'}, usage: str = 'Get answers from Google Search', belief: ~typing.Any = None, output_key: str | None = None, resources: list[~sherpa_ai.actions.base.ActionResource] = <factory>, num_documents: int = 5, reranker: ~sherpa_ai.actions.utils.reranking.BaseReranking = None, refiner: ~sherpa_ai.actions.utils.refinement.BaseRefinement = None, current_task: str = '', perform_reranking: bool = False, perform_refinement: bool = False, role_description: str, task: str, llm: ~typing.Any = None, description: str = 'Role Description: {role_description}\nTask: {task}\n\nRelevant Documents:\n{documents}\n\n\nReview and analyze the provided documents with respect to the task. Craft a concise and short, unified summary that distills key information that is most relevant to the task, incorporating reference links within the summary.\nOnly use the information given. Do not add any additional information. The summary should be less than {n} setences\n', config: ~sherpa_ai.config.task_config.AgentConfig = AgentConfig(verbose=True, gsite=[], do_reflect=False, use_task_agent=False, search_domains=[], invalid_domains=[]))[source]#

Bases: BaseRetrievalAction

args: dict#
belief: Any#
config: AgentConfig#
current_task: str#
description: str#
llm: Any#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

model_post_init(context: Any, /) → None#

This function is meant to behave like a BaseModel method to initialise private attributes.

It takes context as an argument since that’s what pydantic-core passes when calling it.

Parameters:

  • self – The BaseModel instance.

  • context – The context.

name: str#
num_documents: int#
output_key: str | None#
perform_refinement: bool#
perform_reranking: bool#
refiner: BaseRefinement#
reranker: BaseReranking#
resources: list[ActionResource]#
role_description: str#
search(query) → list[dict][source]#

Search for relevant documents based on the query.

task: str#
usage: str#
class sherpa_ai.actions.SynthesizeOutput(*, name: str = 'SynthesizeOutput', args: dict = {'context': 'string', 'history': 'string', 'task': 'string'}, usage: str = 'Answer the question using conversation history with the user', belief: Any = None, output_key: str | None = None, role_description: str, llm: Any = None, description: str = '{role_description}\n\nContext: {context}\n\nAction - Result History:\n{history}\n\nGiven the context and the action-result history, please complete the task mentioned. Include any links you used from the context and history in the result.\nTask: {task}\nResult:\n', add_citation: bool = False)[source]#

Bases: BaseAction

add_citation: bool#
args: dict#
belief: Any#
description: str#
execute(task: str, context: str, history: str) → str[source]#
llm: Any#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str#
output_key: str | None#
role_description: str#
usage: str#
class sherpa_ai.actions.TaskPlanning(*args, name: str = 'TaskPlanning', usage: str = 'Come up with a plan to solve the task', belief: Any = None, output_key: str | None = None, llm: Any = None, num_steps: int = 5, prompt: str = 'You are a **task decomposition assistant** who simplifies complex tasks into sequential steps, assigning roles or agents to each.\nBy analyzing user-defined tasks and agent capabilities, you provides structured plans, enhancing project clarity and efficiency.\nYour adaptability ensures customized solutions for diverse needs.\n\nA good plan is concise, detailed, feasible and efficient.\n\nTask: **{task}**\n\nAgents:\n{agent_pool_description}\n\nPlease break down the task into maximum {num_steps} individual, detailed steps and designate an appropriate agent for each step. The result should be in the following format:\nStep 1:\n    Agent: <AgentName>\n    Task: <detailed task description>\n...\nStep N:\n    Agent: <AgentName>\n    Task: <detailed task description>\n\nDo not answer anything else, and do not add any other information in your answer. Only select agents from the the list and only select one agent at a time.\n', revision_prompt: str = 'You are a **task decomposition assistant** who simplifies complex tasks into sequential steps, assigning roles or agents to each.\nBy analyzing user-defined tasks and agent capabilities, you provide structured plans, enhancing project clarity and efficiency.\nYour adaptability ensures customized solutions for diverse needs.\n\nA good plan is concise, detailed, feasible and efficient. It should be broken down into individual steps, with each step assigned to an appropriate agent.\n\nTask: **{task}**\n\nAgents:\n{agent_pool_description}\n\nHere is your previous plan:\n{previous_plan}\n\nHere is the feedback from the last run:\n{feedback}\n\nPlease revise the plan based on the feedback to maximum {num_steps} steps. The result should be in the following format:\nStep 1:\n    Agent: <AgentName>\n    Task: <detailed task description>\n...\nStep N:\n    Agent: <AgentName>\n    Task: <detailed task description>\n\nDo not answer anything else, and do not add any other information in your answer. Only select agents from the the list and only select one agent at a time.\n')[source]#

Bases: BaseAction

args: dict#
belief: Any#
execute(task: str, agent_pool_description: str, last_plan: str | None = None, feedback: str | None = None) → Plan[source]#

Execute the action

llm: Any#
model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True}#

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str#
num_steps: int#
output_key: str | None#
post_process(action_output: str) → Plan[source]#

Post process the action output into a plan with steps

prompt: str#
revision_prompt: str#
usage: str#
In This Page: