| foundry_agent_create | R Documentation |
Create a named, versioned prompt agent on the project-scoped Agent Service.
The agent bundles a model, system instructions, and optional tools so it can
later be run by name through foundry_response().
foundry_agent_create(
name,
model = NULL,
instructions = NULL,
description = NULL,
metadata = NULL,
temperature = NULL,
top_p = NULL,
tools = NULL,
tool_choice = NULL,
definition = NULL,
api_key = NULL,
token = NULL,
endpoint = NULL,
api_version = "v1"
)
name |
Character. Agent name. Up to 63 characters, alphanumeric and hyphens, unique within the project. |
model |
Character. Model deployment name. Required unless |
instructions |
Character. Optional system prompt. |
description |
Character. Optional human-readable description. |
metadata |
Named list. Optional key-value metadata (up to 16 pairs). |
temperature |
Numeric. Optional sampling temperature in |
top_p |
Numeric. Optional nucleus-sampling value in |
tools |
List. Optional tools: |
tool_choice |
Character or list. Optional tool-choice control. |
definition |
List. Optional full agent definition. When supplied, the
individual |
api_key |
Character. Optional API key. Falls back to configured auth. |
token |
Character. Optional bearer token. Falls back to configured auth. |
endpoint |
Character. Optional project endpoint override. |
api_version |
Character. API version query value. Defaults to |
A one-row tibble describing the created agent.
## Not run:
# Requires a configured Azure project endpoint and credentials,
# plus a model deployment.
foundry_agent_create(
name = "france-facts",
model = "gpt-5-nano",
instructions = "You answer questions about France concisely."
)
## End(Not run)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.