Package index • foundryR

Setup and configuration

Configure your Azure AI Foundry credentials and verify your setup. These functions manage authentication for all API calls.

foundry_check_setup(): Check foundryR Setup
foundry_set_key(): Set Azure AI Foundry API Key
foundry_set_token(): Set Azure AI Foundry Bearer Token
foundry_set_token_provider(): Set a Microsoft Entra ID token provider
foundry_token_azure_cli(): Create an Azure CLI token provider
foundry_token_azure_identity(): Create a Microsoft Entra ID token provider using AzureAuth
foundry_set_endpoint(): Set Azure AI Foundry Endpoint
foundry_get_endpoint(): Get Azure AI Foundry Endpoint
foundry_set_project_endpoint(): Set Azure AI Foundry project endpoint
foundry_get_project_endpoint(): Get Azure AI Foundry project endpoint
foundry_set_speech_endpoint(): Set Microsoft Foundry Speech endpoint
foundry_set_speech_key(): Set Microsoft Foundry Speech API key

Content Safety

Azure AI Content Safety features for responsible AI pipelines. Moderate content, detect hallucinations, and protect against prompt injection.

foundry_moderate(): Moderate Text Content
foundry_moderate_image(): Moderate image content
foundry_moderate_multimodal() experimental: Moderate an image together with its text
foundry_protected_material(): Detect protected material in text
foundry_protected_code() experimental: Detect protected material in code
foundry_blocklists() foundry_blocklist_create() foundry_blocklist_get() foundry_blocklist_delete() foundry_blocklist_items() foundry_blocklist_add_items() foundry_blocklist_remove_items(): Manage Content Safety text blocklists
foundry_groundedness(): Detect Groundedness of LLM Responses
foundry_llm_resource(): Describe a bring-your-own Azure OpenAI resource for groundedness
foundry_shield(): Shield Prompt from Injection Attacks
foundry_task_adherence() experimental: Check an agent transcript for task adherence
foundry_agent_tool(): Describe an agent tool for task adherence
foundry_agent_tool_call(): Describe an agent tool call for task adherence
foundry_agent_message(): Describe an agent message for task adherence
foundry_set_content_safety_key(): Set Azure Content Safety API Key
foundry_set_content_safety_endpoint(): Set Azure Content Safety Endpoint

Responses API and extraction

Use Microsoft Foundry’s newer v1 Responses API for stateful turns, user-defined tools, schema-constrained extraction, and web-grounded answers.

foundry_response(): Create a response with the Azure OpenAI Responses API
foundry_agent(): Run a bounded Responses API tool-calling loop
foundry_tool(): Define an R function as a Responses API tool
foundry_response_retrieve(): Retrieve a stored Responses API response
foundry_response_delete(): Delete a stored Responses API response
foundry_response_cancel(): Cancel a background Responses API response
foundry_response_input_items(): List input items for a Responses API response
foundry_extract(): Extract structured data from text using JSON Schema
foundry_web_search(): Search the web with the Responses API
foundry_schema(): Build a strict JSON Schema object
schema_string() schema_enum() schema_number() schema_integer() schema_boolean() schema_array() schema_object(): Schema constructors for structured outputs
type_boolean() type_enum() type_number() type_string(): Codebook schema helpers
as_foundry_schema(): Convert an object to a foundryR JSON Schema

Agents

Create, list, and manage named, versioned prompt agents on the Foundry Agent Service, then run a stored agent by name through [foundry_response()] with agent_reference.

foundry_agent_create(): Create a Foundry agent
foundry_agents(): List Foundry agents
foundry_agent_get(): Retrieve a Foundry agent
foundry_agent_delete(): Delete a Foundry agent
foundry_agent_versions(): List versions of a Foundry agent
foundry_agent_reference(): Reference a Foundry agent from the Responses API

Conversations and vector stores

Manage server-side Responses API conversations and Azure-hosted vector stores for shared retrieval workflows.

foundry_conversation_create() foundry_conversations() foundry_conversation_get() foundry_conversation_update() foundry_conversation_delete() foundry_conversation_items() foundry_conversation_items_add(): Manage Responses API conversations
foundry_vector_store_create() foundry_vector_stores() foundry_vector_store_get() foundry_vector_store_modify() foundry_vector_store_delete() foundry_vector_store_files() foundry_vector_store_file_add() foundry_vector_store_file_remove() foundry_vector_store_file_batch() foundry_vector_search(): Manage Azure OpenAI vector stores
foundry_tool_file_search(): Create a file-search tool definition

Text embeddings

Generate vector embeddings for semantic search, clustering, and machine learning. Includes batch processing and similarity computation.

foundry_embed(): Generate Text Embeddings
foundry_embed_batch(): Generate Text Embeddings in Parallel Batches
foundry_similarity(): Compute Cosine Similarity Between Embeddings

Files and Batch API

Upload files, prepare JSONL request files, and manage asynchronous batch jobs for large-scale annotation, extraction, and classification.

foundry_file_upload(): Upload a file to Microsoft Foundry
foundry_files(): List uploaded Microsoft Foundry files
foundry_file_get(): Retrieve a Microsoft Foundry file
foundry_file_delete(): Delete a Microsoft Foundry file
foundry_file_download(): Download Microsoft Foundry file content
foundry_batch_requests(): Write JSONL requests for the Batch API
foundry_batch_create(): Create a Microsoft Foundry batch
foundry_batches(): List Microsoft Foundry batches
foundry_batch_get(): Retrieve a Microsoft Foundry batch
foundry_batch_cancel(): Cancel a Microsoft Foundry batch
foundry_batch_wait(): Wait for a Microsoft Foundry batch to finish
foundry_batch_results(): Parse completed Microsoft Foundry batch results
foundry_extract_batch(): Extract structured data with the Batch API
foundry_usage(): Summarise token usage for foundryR results

Evaluations

Run Azure AI Foundry cloud evaluations. Define graders (string check, text similarity, model-graded labels and scores, and Azure built-in evaluators), create evaluations, and inspect per-row grader results as tibbles.

foundry_grader_string_check(): String-check grader
foundry_grader_text_similarity(): Text-similarity grader
foundry_grader_label_model(): Label-model grader
foundry_grader_score_model(): Score-model grader
foundry_grader_azure_ai(): Azure AI built-in evaluator grader
foundry_eval_item(): Build an evaluation item for model-based graders
foundry_eval_data_config(): Define an evaluation data-source configuration
foundry_eval_run_data(): Define an evaluation run data source
foundry_eval_create(): Create an evaluation
foundry_evals(): List evaluations
foundry_eval_get(): Retrieve an evaluation
foundry_eval_delete(): Delete an evaluation
foundry_eval_run_create(): Create an evaluation run
foundry_eval_runs(): List evaluation runs
foundry_eval_run_get(): Retrieve an evaluation run
foundry_eval_run_cancel(): Cancel an evaluation run
foundry_eval_run_output_items(): List evaluation run output items

Validation and provenance

Validate LLM annotation against human labels, measure repeated-extraction consistency, and record reproducibility metadata.

foundry_agreement(): Compute agreement metrics for LLM annotation
foundry_consistency(): Measure repeated-extraction consistency
foundry_codebook(): Create a measurement codebook
codebook_diff(): Compare two codebooks
foundry_provenance(): Capture model and schema provenance

tidymodels integration

Add Foundry text embeddings to tidymodels recipes.

step_foundry_embed() tidy(<step_foundry_embed>): Foundry Embedding Recipe Step
bake(<step_foundry_embed>): Apply the Foundry embedding step to new data
prep(<step_foundry_embed>): Prepare the Foundry embedding step
print(<step_foundry_embed>): Print method for step_foundry_embed
required_pkgs(<step_foundry_embed>): Required packages for step_foundry_embed
foundry_cache_clear(): Clear the foundryR embedding cache

Chat completions

Send chat-completion requests to Azure AI Foundry deployments.

foundry_chat(): Chat with an Azure AI Model

Audio and speech

Transcribe and translate research audio, including MAI-Transcribe models through LLM Speech, and synthesize text-to-speech audio files.

foundry_transcribe(): Transcribe an audio file with Microsoft Foundry
foundry_translate_audio(): Translate an audio file with Microsoft Foundry
foundry_speak() experimental: Generate speech audio from text

Experimental media

Generate and edit images with v1 preview image models, and manage preview video generation jobs.

foundry_image(): Generate Images with DALL-E
foundry_image_edit() experimental: Edit an image with Microsoft Foundry
foundry_save_image(): Save Generated Image to File
foundry_set_image_endpoint(): Set Image Generation Endpoint
foundry_set_image_key(): Set Image Generation API Key
foundry_video_job_create() experimental: Create a Microsoft Foundry video generation job
foundry_video_jobs() experimental: List Microsoft Foundry video generation jobs
foundry_video_job_get() experimental: Retrieve a Microsoft Foundry video generation job
foundry_video_job_delete() experimental: Delete a Microsoft Foundry video generation job
foundry_video_get() experimental: Retrieve a Microsoft Foundry video generation
foundry_video_download() experimental: Download Microsoft Foundry generated video content

Model discovery

Explore available model deployments in your Azure AI Foundry resource.

foundry_models(): List or retrieve available model deployments