PerplexitySearchTool - AG2

PerplexitySearchTool

`autogen.tools.experimental.perplexity.perplexity_search.PerplexitySearchTool`

PerplexitySearchTool(model='sonar', api_key=None, max_tokens=1000, search_domain_filter=None)

Bases: Tool

Tool for interacting with the Perplexity AI search API.

This tool uses the Perplexity API to perform web search, news search, and conversational search, returning concise and precise responses.

ATTRIBUTE	DESCRIPTION
`url`	API endpoint URL. TYPE:`str`
`model`	Name of the model to be used. TYPE:`str`
`api_key`	API key for authenticating with the Perplexity API. TYPE:`str`
`max_tokens`	Maximum tokens allowed for the API response. TYPE:`int`
`search_domain_filters`	Optional list of domain filters for the search. TYPE:`Optional[list[str]]`

Initializes a new instance of the PerplexitySearchTool.

PARAMETER	DESCRIPTION
`model`	The model to use. Defaults to "sonar". TYPE:`str`DEFAULT:`'sonar'`
`api_key`	API key for authentication. TYPE:`Optional[str]`DEFAULT:`None`
`max_tokens`	Maximum number of tokens for the response. Defaults to 1000. TYPE:`int`DEFAULT:`1000`
`search_domain_filter`	List of domain filters to restrict search. TYPE:`Optional[list[str]]`DEFAULT:`None`

RAISES	DESCRIPTION
`ValueError`	If the API key is missing, the model is empty, max_tokens is not positive, or if search_domain_filter is not a list when provided.

Source code in autogen/tools/experimental/perplexity/perplexity_search.py

 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149  def __init__( self, model: str = "sonar", api_key: Optional[str] = None, max_tokens: int = 1000, search_domain_filter: Optional[list[str]] = None, ): """ Initializes a new instance of the PerplexitySearchTool. Args: model (str, optional): The model to use. Defaults to "sonar". api_key (Optional[str], optional): API key for authentication. max_tokens (int, optional): Maximum number of tokens for the response. Defaults to 1000. search_domain_filter (Optional[list[str]], optional): List of domain filters to restrict search. Raises: ValueError: If the API key is missing, the model is empty, max_tokens is not positive, or if search_domain_filter is not a list when provided. """ self.api_key = api_key or os.getenv("PERPLEXITY_API_KEY") self._validate_tool_config(model, self.api_key, max_tokens, search_domain_filter) self.url = "https://api.perplexity.ai/chat/completions" self.model = model self.api_key = api_key self.max_tokens = max_tokens self.search_domain_filters = search_domain_filter super().__init__( name="perplexity-search", description="Perplexity AI search tool for web search, news search, and conversational search " "for finding answers to everyday questions, conducting in-depth research and analysis.", func_or_tool=self.search, )


`<br>117<br>118<br>119<br>120<br>121<br>122<br>123<br>124<br>125<br>126<br>127<br>128<br>129<br>130<br>131<br>132<br>133<br>134<br>135<br>136<br>137<br>138<br>139<br>140<br>141<br>142<br>143<br>144<br>145<br>146<br>147<br>148<br>149<br>`	<br>def __init__(<br> self,<br> model: str = "sonar",<br> api_key: Optional[str] = None,<br> max_tokens: int = 1000,<br> search_domain_filter: Optional[list[str]] = None,<br>):<br> """<br> Initializes a new instance of the PerplexitySearchTool.<br> Args:<br> model (str, optional): The model to use. Defaults to "sonar".<br> api_key (Optional[str], optional): API key for authentication.<br> max_tokens (int, optional): Maximum number of tokens for the response. Defaults to 1000.<br> search_domain_filter (Optional[list[str]], optional): List of domain filters to restrict search.<br> Raises:<br> ValueError: If the API key is missing, the model is empty, max_tokens is not positive,<br> or if search_domain_filter is not a list when provided.<br> """<br> self.api_key = api_key or os.getenv("PERPLEXITY_API_KEY")<br> self._validate_tool_config(model, self.api_key, max_tokens, search_domain_filter)<br> self.url = "https://api.perplexity.ai/chat/completions"<br> self.model = model<br> self.api_key = api_key<br> self.max_tokens = max_tokens<br> self.search_domain_filters = search_domain_filter<br> super().__init__(<br> name="perplexity-search",<br> description="Perplexity AI search tool for web search, news search, and conversational search "<br> "for finding answers to everyday questions, conducting in-depth research and analysis.",<br> func_or_tool=self.search,<br> )<br>

`url` instance-attribute

url = 'https://api.perplexity.ai/chat/completions'

`model` instance-attribute

model = model

`api_key` instance-attribute

api_key = api_key

`max_tokens` instance-attribute

max_tokens = max_tokens

`search_domain_filters` instance-attribute

search_domain_filters = search_domain_filter

`name` property

name

`description` property

description

`func` property

func

`tool_schema` property

tool_schema

Get the schema for the tool.

This is the preferred way of handling function calls with OpenAI and compatible frameworks.

`function_schema` property

function_schema

Get the schema for the function.

This is the old way of handling function calls with OpenAI and compatible frameworks. It is provided for backward compatibility.

`realtime_tool_schema` property

realtime_tool_schema

Get the schema for the tool.

This is the preferred way of handling function calls with OpenAI and compatible frameworks.

`search`

search(query)

Perform a search query using the Perplexity AI API.

Constructs the payload, executes the query, and parses the response to return a concise search result along with any provided citations.

PARAMETER	DESCRIPTION
`query`	The search query. TYPE:`str`

RETURNS	DESCRIPTION
`SearchResponse`	A model containing the search result content and citations. TYPE:`SearchResponse`

RAISES	DESCRIPTION
`ValueError`	If the search query is invalid.
`RuntimeError`	If there is an error during the search process.

Source code in autogen/tools/experimental/perplexity/perplexity_search.py

 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225  def search(self, query: Annotated[str, "The search query."]) -> SearchResponse: """ Perform a search query using the Perplexity AI API. Constructs the payload, executes the query, and parses the response to return a concise search result along with any provided citations. Args: query (str): The search query. Returns: SearchResponse: A model containing the search result content and citations. Raises: ValueError: If the search query is invalid. RuntimeError: If there is an error during the search process. """ payload = { "model": self.model, "messages": [{"role": "system", "content": "Be precise and concise."}, {"role": "user", "content": query}], "max_tokens": self.max_tokens, "search_domain_filter": self.search_domain_filters, "web_search_options": {"search_context_size": "high"}, } try: perplexity_response = self._execute_query(payload) content = perplexity_response.choices[0].message.content citations = perplexity_response.citations return SearchResponse(content=content, citations=citations, error=None) except Exception as e: return SearchResponse( content=None, citations=None, error=f"PerplexitySearchTool failed to search. Error: {e}" )


`<br>193<br>194<br>195<br>196<br>197<br>198<br>199<br>200<br>201<br>202<br>203<br>204<br>205<br>206<br>207<br>208<br>209<br>210<br>211<br>212<br>213<br>214<br>215<br>216<br>217<br>218<br>219<br>220<br>221<br>222<br>223<br>224<br>225<br>`	<br>def search(self, query: Annotated[str, "The search query."]) -> SearchResponse:<br> """<br> Perform a search query using the Perplexity AI API.<br> Constructs the payload, executes the query, and parses the response to return<br> a concise search result along with any provided citations.<br> Args:<br> query (str): The search query.<br> Returns:<br> SearchResponse: A model containing the search result content and citations.<br> Raises:<br> ValueError: If the search query is invalid.<br> RuntimeError: If there is an error during the search process.<br> """<br> payload = {<br> "model": self.model,<br> "messages": [{"role": "system", "content": "Be precise and concise."}, {"role": "user", "content": query}],<br> "max_tokens": self.max_tokens,<br> "search_domain_filter": self.search_domain_filters,<br> "web_search_options": {"search_context_size": "high"},<br> }<br> try:<br> perplexity_response = self._execute_query(payload)<br> content = perplexity_response.choices[0].message.content<br> citations = perplexity_response.citations<br> return SearchResponse(content=content, citations=citations, error=None)<br> except Exception as e:<br> return SearchResponse(<br> content=None, citations=None, error=f"PerplexitySearchTool failed to search. Error: {e}"<br> )<br>

`register_for_llm`

register_for_llm(agent)

Registers the tool for use with a ConversableAgent's language model (LLM).

This method registers the tool so that it can be invoked by the agent during interactions with the language model.

PARAMETER	DESCRIPTION
`agent`	The agent to which the tool will be registered. TYPE:`ConversableAgent`

Source code in autogen/tools/tool.py

 72 73 74 75 76 77 78 79 80 81   def register_for_llm(self, agent: "ConversableAgent") -> None: """Registers the tool for use with a ConversableAgent's language model (LLM). This method registers the tool so that it can be invoked by the agent during interactions with the language model. Args: agent (ConversableAgent): The agent to which the tool will be registered. """ agent.register_for_llm()(self) 


`<br>72<br>73<br>74<br>75<br>76<br>77<br>78<br>79<br>80<br>81<br>`	`<br>def register_for_llm(self, agent: "ConversableAgent") -> None:<br> """Registers the tool for use with a ConversableAgent's language model (LLM).<br> This method registers the tool so that it can be invoked by the agent during<br> interactions with the language model.<br> Args:<br> agent (ConversableAgent): The agent to which the tool will be registered.<br> """<br> agent.register_for_llm()(self)<br>`

`register_for_execution`

register_for_execution(agent)

Registers the tool for direct execution by a ConversableAgent.

This method registers the tool so that it can be executed by the agent, typically outside of the context of an LLM interaction.

Source code in autogen/tools/tool.py

 83 84 85 86 87 88 89 90 91 92   def register_for_execution(self, agent: "ConversableAgent") -> None: """Registers the tool for direct execution by a ConversableAgent. This method registers the tool so that it can be executed by the agent, typically outside of the context of an LLM interaction. Args: agent (ConversableAgent): The agent to which the tool will be registered. """ agent.register_for_execution()(self) 


`<br>83<br>84<br>85<br>86<br>87<br>88<br>89<br>90<br>91<br>92<br>`	`<br>def register_for_execution(self, agent: "ConversableAgent") -> None:<br> """Registers the tool for direct execution by a ConversableAgent.<br> This method registers the tool so that it can be executed by the agent,<br> typically outside of the context of an LLM interaction.<br> Args:<br> agent (ConversableAgent): The agent to which the tool will be registered.<br> """<br> agent.register_for_execution()(self)<br>`

`register_tool`

register_tool(agent)

Equivalent to calling both register_for_llm and register_for_execution with the same agent.

Note: This will not make the agent recommend and execute the call in the one step. If the agent recommends the tool, it will need to be the next agent to speak in order to execute the tool.

Source code in autogen/tools/tool.py

 94 95 96 97 98 99 100 101 102 103 104 105 106  def register_tool(self, agent: "ConversableAgent") -> None: """Register a tool to be both proposed and executed by an agent. Equivalent to calling both `register_for_llm` and `register_for_execution` with the same agent. Note: This will not make the agent recommend and execute the call in the one step. If the agent recommends the tool, it will need to be the next agent to speak in order to execute the tool. Args: agent (ConversableAgent): The agent to which the tool will be registered. """ self.register_for_llm(agent) self.register_for_execution(agent)


`<br> 94<br> 95<br> 96<br> 97<br> 98<br> 99<br>100<br>101<br>102<br>103<br>104<br>105<br>106<br>`	<br>def register_tool(self, agent: "ConversableAgent") -> None:<br> """Register a tool to be both proposed and executed by an agent.<br> Equivalent to calling both `register_for_llm` and `register_for_execution` with the same agent.<br> Note: This will not make the agent recommend and execute the call in the one step. If the agent<br> recommends the tool, it will need to be the next agent to speak in order to execute the tool.<br> Args:<br> agent (ConversableAgent): The agent to which the tool will be registered.<br> """<br> self.register_for_llm(agent)<br> self.register_for_execution(agent)<br>

PerplexitySearchTool

autogen.tools.experimental.perplexity.perplexity_search.PerplexitySearchTool

url instance-attribute

model instance-attribute

api_key instance-attribute

max_tokens instance-attribute

search_domain_filters instance-attribute

name property

description property

func property

tool_schema property

function_schema property

realtime_tool_schema property

search

register_for_llm

register_for_execution

register_tool

`autogen.tools.experimental.perplexity.perplexity_search.PerplexitySearchTool`

`url` instance-attribute

`model` instance-attribute

`api_key` instance-attribute

`max_tokens` instance-attribute

`search_domain_filters` instance-attribute

`name` property

`description` property

`func` property

`tool_schema` property

`function_schema` property

`realtime_tool_schema` property

`search`

`register_for_llm`

`register_for_execution`

`register_tool`