---
title: "Reproducible Agentic Pipelines with targets"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Reproducible Agentic Pipelines with targets}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  eval = FALSE
)
library(HydraR)
```

`HydraR` is designed to be a building block for complex, long-running scientific workflows. When combined with the [targets](https://books.ropensci.org/targets/) package, you can create reproducible agentic pipelines that cache expensive LLM calls and manage complex dependencies.

## The "Agent-as-Target" Pattern

The most straightforward way to integrate `HydraR` is to encapsulate an `AgentDAG` run within a discrete target. This ensures that the agent only runs if its inputs (data, prompts, or code) change.

### Example: Literature Synthesis Pipeline

Imagine a pipeline where we first fetch raw data, and then an agent summarizes it. We define the agent's logic in `targets_integration.yml`.

```r
library(targets)
library(HydraR)

# _targets.R
list(
  tar_target(
    raw_document,
    "Highly complex scientific text about bioinformatics trends in 2026..."
  ),
  
  tar_target(
    agent_results,
    {
      # 1. Load the workflow definition from YAML
      wf <- load_workflow("targets_integration.yml")
      
      # 2. Spawn the DAG
      dag <- spawn_dag(wf)
      
      # 3. Run with data from 'raw_document'
      dag$run(initial_state = list(raw_text = raw_document))
    }
  ),
  
  tar_target(
    final_summary,
    agent_results$state$get("summarizer")
  )
)
```

## Benefits of Integration

### 1. Caching Expensive LLM Calls
LLM interactions can be costly and slow. By wrapping a `HydraR` run in a target, `targets` will skip the entire agentic workflow if the inputs haven't changed, saving you tokens and time.

### 2. Dependency Tracking
If you modify a system prompt or a tool function that your agent uses, `targets` will detect the change and re-run the agent automatically. This ensures your scientific results are always in sync with your latest configurations.

### 3. Resilience and Resumption
`HydraR` nodes can be configured to use a `Checkpointer`. While `targets` manages the high-level cache, `HydraR` can manage the internal state of the agent. If a large DAG fails at step 50, you can resume from the `HydraR` checkpoint within the same target execution.

## Conclusion

Combining `HydraR` and `targets` brings professional software engineering rigor to agentic R workflows. It transforms "experimental scripts" into robust, auditable, and reproducible scientific pipelines.

---
<!-- APAF Bioinformatics | targets_integration.Rmd | Approved | 2026-04-03 -->