haive.agents.research.open_perplexity.state¶

State schemas for the open_perplexity research agent.

This module defines the state schemas used by the research agent to track the progress of research, manage search queries, store sources, and generate reports. It includes schemas for input, processing state, and output.

Classes¶

`ReportSection`	Represents a section in the research report.
`ResearchConfidenceLevel`	Confidence level in research findings.
`ResearchInputState`	Input state for the research process.
`ResearchOutputState`	Output state for the research process.
`ResearchState`	State schema for the open_perplexity research agent.
`WebSearchQuery`	Represents a web search query for retrieving information.

Module Contents¶

class haive.agents.research.open_perplexity.state.ReportSection(/, **data)¶

Bases: pydantic.BaseModel

Represents a section in the research report.

Tracks a single report section, including its content, research needs, and associated queries and sources.

Parameters:: data (Any)

name¶: Section name

description¶: Section description

content¶: Section content

requires_research¶: Whether this section requires research

queries¶: Search queries for this section

sources¶: Sources used in this section

status¶: Section status (pending, in_progress, completed)

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

class haive.agents.research.open_perplexity.state.ResearchConfidenceLevel¶

Bases: str, enum.Enum

Confidence level in research findings.

Represents the overall confidence level in the research results, based on source quality, quantity, and consistency.

HIGH¶: High confidence based on numerous reliable sources

MEDIUM¶: Medium confidence with good but limited sources

LOW¶: Low confidence due to limited or questionable sources

INSUFFICIENT_DATA¶: Not enough data to establish confidence

Initialize self. See help(type(self)) for accurate signature.

class haive.agents.research.open_perplexity.state.ResearchInputState(/, **data)¶

Bases: pydantic.BaseModel

Input state for the research process.

Represents the initial input to the research agent, including the user’s query and any additional context.

Parameters:: data (Any)

messages¶: Input messages including user query

input_context¶: Additional context provided for the research

research_parameters¶: Optional parameters to customize the research process

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

class haive.agents.research.open_perplexity.state.ResearchOutputState(/, **data)¶

Bases: pydantic.BaseModel

Output state for the research process.

Represents the final output from the research agent, including the completed report and confidence assessment.

Parameters:: data (Any)

final_report¶: Complete research report in markdown format

confidence_level¶: Confidence level in the research findings

sources¶: Sources used in the research

messages¶: Conversation history including the assistant’s final response

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

class haive.agents.research.open_perplexity.state.ResearchState(/, **data)¶

Bases: pydantic.BaseModel

State schema for the open_perplexity research agent.

Comprehensive state object that tracks all aspects of the research process, including messages, research parameters, report sections, sources, and findings.

Parameters:: data (Any)

messages¶: Conversation messages

research_topic¶: Main topic of research

research_question¶: Specific research question

input_context¶: Additional context provided by the user

search_parameters¶: Parameters for search and research customization

report_sections¶: Sections of the research report

current_section_index¶: Index of the current section being researched

search_queries¶: List of search queries to execute

query¶: Current search query

retrieved_documents¶: Documents retrieved from search

sources¶: Sources used in the research

data_sources¶: Data sources available and used

vectorstore_documents¶: Documents loaded into vector store

research_findings¶: Key findings from the research

confidence_level¶: Overall confidence level in research findings

confidence_explanation¶: Explanation for the confidence level assessment

final_report¶: Final research report content

current_step¶: Current step in the workflow

error¶: Error message if any step fails

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

class haive.agents.research.open_perplexity.state.WebSearchQuery(/, **data)¶

Bases: pydantic.BaseModel

Represents a web search query for retrieving information.

Tracks a single search query, its purpose, execution status, and results.

Parameters:: data (Any)

query¶: The search query text

purpose¶: Purpose of this search query

data_source¶: Data source to query (web, github, academic, news, etc.)

completed¶: Whether this query has been executed

results¶: Search results

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.