Struct AgentSource 

pub struct AgentSource { /* private fields */ }

Adapter: LLM agent as a pipeline data source.

Wraps any AIProvider and exposes it through AgentSourcePort and ScrapingService for integration into DAG pipelines.

Implementations

impl AgentSource

pub fn new(provider: Arc<dyn AIProvider>) -> Self

Create a new agent source backed by the given AI provider.

Arguments

• provider - An Arc-wrapped AIProvider implementation.

Example

use stygian_graph::adapters::agent_source::AgentSource;
use stygian_graph::adapters::mock_ai::MockAIProvider;
use std::sync::Arc;

let source = AgentSource::new(Arc::new(MockAIProvider));

Trait Implementations

impl AgentSourcePort for AgentSource

fn invoke<'life0, 'async_trait>( &'life0 self, request: AgentRequest, ) -> Pin<Box<dyn Future<Output = Result<AgentResponse>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Invoke the agent with the given request.

fn source_name(&self) -> &str

Name of this agent source for logging and identification.

impl ScrapingService for AgentSource

fn execute<'life0, 'async_trait>( &'life0 self, input: ServiceInput, ) -> Pin<Box<dyn Future<Output = Result<ServiceOutput>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Invoke the LLM agent with prompt data from the pipeline.

Expected params:

{ "prompt": "Summarise this page", "parameters": {} }

The input.url field is ignored; the prompt and optional upstream data are passed via params.

fn name(&self) -> &'static str

Service name for identification in logs and metrics