Skip to contents

Create a PRA Risk Analysis Chat Agent

Source: R/chat.R
pra_chat.Rd

Creates an ellmer chat object configured as a project risk analysis expert, with all PRA functions registered as tools and optional RAG context retrieval from the bundled knowledge base.

Usage

pra_chat(
  chat = NULL,
  model = "llama3.2",
  rag = TRUE,
  embed_model = .pra_default_embed_model
)

Arguments

chat

An optional pre-configured ellmer chat object. If provided, model is ignored and tools are registered on this object instead.

model

Character. Ollama model name (default "llama3.2"). Must support tool calling. Other options: "qwen2.5", "llama3.1:70b".

rag

Logical. Whether to use RAG context from the PRA knowledge base (default TRUE). Requires Ollama embedding model to be available.

embed_model

Character. Ollama embedding model for RAG (default "nomic-embed-text"). Only used when rag = TRUE.

Value

A configured ellmer chat object with PRA tools registered. Use chat$chat("your question") to interact.

Details

By default, uses a local Ollama model for fully offline, private operation. Alternatively, supply a pre-configured ellmer chat object (e.g., ellmer::chat_openai()) via the chat parameter for cloud-hosted models.

Examples

if (FALSE) { # \dontrun{
# Default: local Ollama model
chat <- pra_chat()
chat$chat("Run a Monte Carlo simulation for a 3-task project with
  Task A: normal(10, 2), Task B: triangular(5, 10, 15), Task C: uniform(8, 12)")

# Use a cloud model for better accuracy
chat <- pra_chat(chat = ellmer::chat_openai(model = "gpt-4o"))

# Follow-up questions use conversation context
chat$chat("What is the contingency reserve at 95% confidence?")
} # }