Explore available tools and example prompts for Foundry MCP Server (preview)

Reference guide for all Foundry MCP Server tools, including tool descriptions, key inputs, permissions, and example prompts for each tool.

This reference documents 79 Foundry MCP Server tools across 17 categories that let you manage agents, sessions, toolboxes, datasets, evaluations, model deployments, continuous evaluation, and more — all through conversational prompts instead of API calls. Use it to explore each tool and try the example prompts in your own project.

Before using these tools, complete the Foundry MCP Server setup.

This feature is currently in public preview. This preview is provided without a service-level agreement, and we don’t recommend it for production workloads. Certain features might not be supported or might have constrained capabilities. For more information, see Supplemental Terms of Use for Microsoft Azure Previews.

How tools work

When you type a natural-language prompt in an MCP-compliant client (for example, GitHub Copilot Agent Mode), the language model selects the appropriate tool and formulates the required parameters on your behalf. You don’t call tools directly — you describe what you want, and the model translates your intent into a tool call.

Each tool is classified as read (retrieves information) or write (creates, updates, or deletes resources). Write operations affect live resources and billing immediately. Review the security best practices before running write operations.

Permissions

All operations run with the authenticated user’s permissions through the Microsoft Entra ID On-Behalf-Of flow. You need the following roles:

Operation type	Minimum Azure role	Notes
Read tools	Reader on the Foundry project or account	Sufficient for listing, querying, and monitoring.
Write tools	Contributor on the Foundry project or account	Required for creating, updating, and deleting resources.
Conditional Access admin	Conditional Access Administrator in Entra ID	Only needed if configuring tenant-level access policies.

For more information, see Role-based access control for Microsoft Foundry.

Key identifiers

Many tools require resource identifiers. The language model extracts these from your prompt context, but it helps to know the formats:

Identifier	Format	Where to find it
Foundry resource ID	/subscriptions/{sub_id}/resourceGroups/{rg}/providers/Microsoft.CognitiveServices/accounts/{account}	Azure portal Properties page
Project endpoint	https://{account}.services.ai.azure.com/api/projects/{project}	Foundry project details page
Project resource ID	/subscriptions/{sub_id}/resourceGroups/{rg}/providers/Microsoft.CognitiveServices/accounts/{account}/projects/{project}	Azure portal Properties page

Agent management

Manage the full lifecycle of agents in a Foundry project, including creation, invocation, container orchestration, and deletion.

Example prompts:

• “List all agents in my Foundry project.”
• “Create a new agent named faq-agent using model gpt-4o-mini.”
• “Send ‘Hello, how can you help?’ to my customer-support-agent.”
• “Start the container for my hosted agent triage-agent.”
• “Check the container status for triage-agent.”
• “Show me the agent definition schema for prompt agents.”
• “Delete the old-test-agent from my project.”

Tool	Access	Description	Key inputs	Returns
agent_get	read	List all agents in a Foundry project, or get a specific agent by name.	Agent name (optional)	Agent list or single agent definition with model, instructions, and tool configuration.
agent_update	write	Create, update, or clone an agent. Use agent_definition_schema_get to discover the full definition schema first.	Agent name, model, instructions, tool definitions	Created or updated agent definition.
agent_invoke	write	Send a message to an agent and get a response. Works for both prompt-based and hosted container agents.	Agent name, message text	Agent response message.
agent_delete	write	Permanently delete an agent. For hosted agents, this also deletes the container.	Agent name	Deletion confirmation.
agent_container_control	write	Start or stop a hosted agent container. Use before invoking a hosted agent.	Agent name, action (start or stop)	Container operation status.
agent_container_status_get	read	Check the current status of a hosted agent container (Starting, Running, Stopped, Failed, and so on).	Agent name	Current container status.
agent_definition_schema_get	read	Return the complete JSON schema for agent definitions, including all tool types.	None	Full JSON schema for agent definitions.

Hosted agent sessions and files

Create and manage isolated compute sessions for hosted agents. Inspect logs and manage files in each session’s filesystem.

Example prompts:

• “Create a session for my hosted agent data-analyst.”
• “List the sessions for data-analyst and show the newest first.”
• “Stream up to 100 log lines from session analysis-2026.”
• “Upload input.csv to /data/input.csv in the session.”
• “Download /data/results.csv from the session.”
• “Delete session analysis-2026 and release its compute resources.”

Tool	Access	Description	Key inputs	Returns
session_create	write	Create a new session for a hosted agent. Sessions provide isolated compute environments for agent execution.	Agent name, session ID (optional)	Created session with its ID, status, and expiration.
session_get	read	Get details for a specific hosted agent session.	Agent name, session ID	Session status, version, creation time, and expiration.
session_list	read	List sessions for a hosted agent with pagination and sorting.	Agent name, limit, order, and after or before pagination cursor	Session list.
session_delete	write	Delete a hosted agent session and release its compute resources.	Agent name, session ID	Operation result.
session_logstream	read	Stream console output from a hosted agent session. The response is capped to prevent unbounded reads.	Agent name, session ID, maximum lines (optional)	Structured standard output and error log events.
session_file_list	read	List files and directories at a path in a hosted agent session.	Agent name, session ID, path (optional)	File and directory list.
session_file_stat	read	Get metadata for a file or directory in a hosted agent session.	Agent name, session ID, file path	Name, size, directory indicator, and last modified time.
session_file_mkdir	write	Create a directory in a hosted agent session.	Agent name, session ID, path, create-parents flag (optional), permission mode (optional)	Operation result.
session_file_upload	write	Upload base64-encoded content to a file in a hosted agent session.	Agent name, session ID, destination file path, base64 content	Operation result.
session_file_download	read	Download a file from a hosted agent session.	Agent name, session ID, file path	Base64-encoded file content.
session_file_delete	write	Delete a file or directory from a hosted agent session.	Agent name, session ID, file path, recursive flag (optional)	Operation result.

Toolbox management

Create, retrieve, version, update, and delete toolboxes in a Foundry project.

Example prompts:

• “Show me the customer-support-tools toolbox.”
• “Get version 2 of customer-support-tools.”
• “Create a new version of customer-support-tools.”
• “Set version 2 of customer-support-tools as the default.”
• “Set version 1 of customer-support-tools as the default, then delete version 2.”
• “Delete the old-support-tools toolbox.”

Toolbox versions are immutable. toolbox_version_create also creates the toolbox if it doesn’t exist. For an existing toolbox, creating a version doesn’t change the default version. Call toolbox_update with defaultVersion set to the new version to promote it. Before you delete the current default version, set another version as the default.

Tool	Access	Description	Key inputs	Returns
toolbox_get	read	Retrieve a toolbox and its current default version.	Toolbox identifier	Toolbox details.
toolbox_version_get	read	List toolbox versions, or retrieve a specific version.	Toolbox identifier, version identifier for a specific version	Toolbox version list or details.
toolbox_version_create	write	Create an immutable toolbox version. If the toolbox doesn’t exist, this tool also creates it.	Toolbox identifier and version details	Created toolbox version.
toolbox_update	write	Create or update a toolbox, including its default version.	Toolbox identifier, updated toolbox details, defaultVersion when changing the default	Created or updated toolbox details.
toolbox_delete	write	Delete a toolbox.	Toolbox identifier	Deletion confirmation.
toolbox_version_delete	write	Delete a specific toolbox version. To delete the current default, first set another version as the default.	Toolbox identifier, version identifier	Deletion confirmation.

Dataset management

Create, retrieve, version, and download evaluation datasets in a Foundry project.

Example prompts:

• “Upload my customer support Q&A dataset from this Azure Blob Storage URL.”
• “Show me all datasets in my Foundry project.”
• “Get details for the customer-support-qa dataset version 2.”
• “List all versions of my product-reviews dataset.”
• “Get a download URL for the latest version of customer-support-qa.”

Tool	Access	Description	Key inputs	Returns
evaluation_dataset_create	write	Create or update a dataset version from an Azure Blob Storage URI.	Dataset name, version, Blob Storage URI	Dataset metadata with name, version, and URI.
evaluation_dataset_get	read	Get a dataset by name and version, or list all datasets in the project.	Dataset name and version (optional)	Dataset details or list of all datasets.
evaluation_dataset_versions_get	read	List all versions of a specific dataset.	Dataset name	List of version numbers with metadata.
evaluation_dataset_sas_url_get	read	Get a short-lived shared access signature (SAS) URL for downloading a dataset version’s content. The version defaults to the latest.	Dataset name, dataset version (optional)	Short-lived dataset content URL.

Data generation jobs

Create and manage standalone data generation jobs for evaluation or fine-tuning data. Generate data from prompts, agents, traces, datasets, or uploaded files.

Example prompts:

• “Generate 100 evaluation question-and-answer samples from this prompt.”
• “Show me the status of data generation job job-123.”
• “Cancel data generation job job-123.”
• “Delete the completed data generation job job-123.”

Tool	Access	Description	Key inputs	Returns
data_generation_job_create	write	Start a standalone data generation job for evaluation or fine-tuning data.	Job name, generation type, scenario, maximum samples, generation model deployment, and source inputs	Operation result.
data_generation_job_get	read	Get one data generation job by ID, or list data generation jobs when the ID is omitted.	Job ID (optional)	Job details or job list.
data_generation_job_cancel	write	Cancel a data generation job.	Job ID	Operation result.
data_generation_job_delete	write	Delete a data generation job.	Job ID	Operation result.

Evaluation operations

Run batch evaluations against agents or datasets, and compare results across runs.

Example prompts:

• “Evaluate my customer-support-agent v2 using Relevance, Groundedness, and Coherence evaluators.”
• “Run a batch evaluation on my JSONL dataset with Violence and HateUnfairness evaluators.”
• “Generate 50 synthetic test queries and evaluate my agent with them.”
• “Show me all evaluation runs in my Foundry project.”
• “Compare run-baseline-123 against treatment runs run-124 and run-125.”

Tool	Access	Description	Key inputs	Returns
evaluation_agent_batch_eval_create	write	Create a batch evaluation run that calls a specific agent. Supports built-in and custom evaluators, plus synthetic data generation.	Agent name/version, evaluator names, dataset (optional for synthetic generation), number of synthetic queries (optional)	Evaluation run ID and status.
evaluation_dataset_batch_eval_create	write	Create a batch evaluation run against a JSONL dataset. Supports built-in and custom evaluators.	Dataset name/version, evaluator names	Evaluation run ID and status.
evaluation_get	read	List evaluation groups or runs in the Foundry project. Set isRequestForRuns to true to list runs within a specific evaluation group.	Evaluation group ID (optional), evaluation run ID (optional), isRequestForRuns flag	List of evaluation groups or runs with status and scores, or details for a specific item.
evaluation_comparison_create	write	Create comparison results between a baseline and treatment evaluation runs.	Baseline run ID, treatment run IDs	Comparison insight ID.
evaluation_comparison_get	read	Get or list evaluation comparison insights.	Comparison insight ID (optional)	Comparison results with statistical analysis.

Trace evaluation

Run multi-turn conversation evaluations over agent traces or explicitly selected conversations and W3C trace IDs.

Example prompts:

• “Evaluate up to five traces from customer-support-agent from the last 24 hours for groundedness and task completion.”
• “Evaluate these conversation IDs for customer satisfaction at the conversation level.”
• “Evaluate these W3C trace IDs with my custom tone-check evaluator.”

Tool	Access	Description	Key inputs	Returns
evaluation_agent_traces_batch_eval_create	write	Create a multi-turn batch evaluation over recent traces sampled from an agent. Creates an evaluation group when one isn’t provided.	Evaluator names, agent name and version or agent ID, evaluation ID (optional), time range or lookback, maximum traces	Evaluation run.
evaluation_traces_batch_eval_create	write	Create a multi-turn batch evaluation over trace data selected by conversation IDs or W3C trace IDs. Creates an evaluation group when one isn’t provided.	Evaluator names, trace source type, conversation IDs or trace IDs, evaluation ID (optional), evaluation level (optional)	Evaluation run.

Evaluation suites

Create, version, run, update, and delete evaluation suites that group datasets, target agents, and evaluators. You can also generate suites from prompts, agents, traces, datasets, or uploaded files.

Example prompts:

• “Create an evaluation suite named support-quality for my support dataset and agent.”
• “Run version 2 of support-quality.”
• “Update the display name and tags for version 2 of support-quality.”
• “Generate an evaluation suite from recent traces for customer-support-agent.”
• “Show me all evaluation suite generation jobs.”

Tool	Access	Description	Key inputs	Returns
evaluation_suite_create	write	Create an evaluation suite version that groups evaluators and can reference a dataset and target agent.	Suite name, evaluator names, dataset and target agent (optional), evaluation settings (optional)	Evaluation suite version.
evaluation_suite_get	read	Get an evaluation suite version, list versions for a suite, or list suites.	Suite name (optional), version (optional), listVersions flag, and agent filter (optional)	Suite details, version list, or suite list.
evaluation_suite_update	write	Update the display name, description, or tags on an evaluation suite version.	Suite name, version, fields to update	Operation result.
evaluation_suite_run	write	Start an evaluation suite run.	Suite name, version (optional), evaluation name (optional), evaluation level (optional)	Evaluation run.
evaluation_suite_delete	write	Delete an evaluation suite version. The version defaults to the latest.	Suite name, version (optional)	Operation result.
evaluation_suite_generation_job_create	write	Start an evaluation suite generation job.	Suite name, generation model deployment, source inputs, generation settings	Generation job.
evaluation_suite_generation_job_get	read	Get one evaluation suite generation job by ID, or list jobs when the ID is omitted.	Job ID (optional), limit and continuation token (optional)	Generation job details or job list.
evaluation_suite_generation_job_cancel	write	Cancel an evaluation suite generation job.	Job ID	Operation result.
evaluation_suite_generation_job_delete	write	Delete an evaluation suite generation job.	Job ID	Operation result.

Evaluator catalog

Browse built-in evaluators and manage custom evaluators for use in evaluation runs.

Example prompts:

• “List all built-in evaluators available in my project.”
• “Show me the full definition of the coherence evaluator.”
• “Create a custom prompt-based evaluator called tone-check that scores responses on a 1-5 scale.”
• “Update the description of my tone-check evaluator.”
• “Delete version 1 of my old-evaluator.”

Tool	Access	Description	Key inputs	Returns
evaluator_catalog_get	read	List evaluators in the catalog, or get the full definition of a specific evaluator. Filter by built-in or custom type.	Evaluator name (optional), type filter (built-in or custom, optional)	Evaluator list or full evaluator definition with scoring logic.
evaluator_catalog_create	write	Create a custom prompt-based or code-based evaluator.	Evaluator name, type (prompt or code), definition	Created evaluator metadata.
evaluator_catalog_update	write	Update metadata (display name, description, category) for an existing custom evaluator.	Evaluator name, fields to update	Updated evaluator metadata.
evaluator_catalog_delete	write	Delete a specific version of a custom evaluator.	Evaluator name, version	Deletion confirmation.

Evaluator generation jobs

Generate evaluators adaptively from prompts, agents, or datasets, and monitor or cancel the generation jobs.

Example prompts:

• “Generate an evaluator named policy-compliance from this policy prompt.”
• “Regenerate policy-compliance from version 2 of my support dataset using model deployment gpt-4o.”
• “Show me all running evaluator generation jobs.”
• “Cancel evaluator generation job job-456.”

Tool	Access	Description	Key inputs	Returns
evaluator_generation_job_create	write	Start an adaptive evaluator generation job. Supplying an existing evaluator name regenerates it from updated source inputs.	Model deployment, evaluator name, prompt, agent, or dataset source inputs	Generation job.
evaluator_generation_job_get	read	Get one evaluator generation job by ID, or list jobs when the ID is omitted.	Job ID (optional), status and agent filters (optional), pagination inputs (optional)	Generation job details or job list.
evaluator_generation_job_cancel	write	Cancel an evaluator generation job.	Job ID	Operation result.

Model catalog and details

Explore models in the Foundry model catalog, get model details, and determine whether a model can be deployed to Managed Compute.

Example prompts:

• “Show me all GPT-5.4 models available in the catalog.”
• “List all Microsoft-published models with MIT license.”
• “Get detailed information and code samples for GPT-5-mini.”
• “Show me models advertised for Managed Compute.”
• “Get Managed Compute details and deployment templates for openai--gpt-oss-20b.”

For Managed Compute, isManagedCompute in model_catalog_list identifies an advertised candidate. Confirm deployability with model_details_get: isFoundryManagedCompute is the authoritative signal, and the resolved deployment templates identify the available deployment options and accelerators.

Tool	Access	Description	Key inputs	Returns
model_catalog_list	read	List models from the Foundry model catalog with optional filters. For Managed Compute candidates, the response includes isManagedCompute.	Search keywords, publisher, license type, task type (all optional)	List of models with name, publisher, license, capabilities, and catalog deployment indicators.
model_details_get	read	Get full model details and code samples. For Managed Compute, use isFoundryManagedCompute to confirm deployability.	Model name or ID	Model specifications, pricing, supported regions, code samples, and resolved deployment templates.

Model deployment management

Deploy, inspect, and remove model deployments, including Managed Compute deployments, in a Foundry account.

Example prompts:

• “Use model_deploy_azure_direct_model to deploy GPT-5-mini as production-chatbot with 20 capacity units.”
• “Show me all my current model deployments.”
• “Show me all my Managed Compute deployments.”
• “Use model_deploy_managed_compute to deploy openai--gpt-oss-20b as gpt-oss-managed using its A100_80GB deployment template.”
• “Delete the old-test-deployment that I’m no longer using.”

Choose explicit tools for each deployment type:

Deployment type	Create or update
Foundry Models sold by Azure	model_deploy_azure_direct_model
Managed Compute	model_deploy_managed_compute

Don’t use the deprecated model_deploy alias for new calls. For Managed Compute, first use model_catalog_list as a candidate signal, then call model_details_get. Use model_deploy_managed_compute only when isFoundryManagedCompute is true, and pass one of the resolved deployment templates. Use model_deployment_get to retrieve either deployment type, and model_deployment_delete to delete either type.

Tool	Access	Description	Key inputs	Returns
model_deploy_azure_direct_model	write	Create or update a deployment for a Foundry Model sold by Azure.	Model name, deployment name, capacity units	Deployment details with endpoint and provisioned capacity.
model_deploy	write	Deprecated backward-compatible alias of model_deploy_azure_direct_model. It has identical inputs and behavior.	Model name, deployment name, capacity units	Deployment details with endpoint and provisioned capacity.
model_deploy_managed_compute	write	Create a distinct Managed Compute deployment from a registry model asset and resolved deployment template.	Deployment name, modelAssetId, deploymentTemplate, acceleratorType (optional), instanceCount (optional), versionUpgradeOption (optional)	Managed Compute deployment details.
model_deployment_get	read	Get deployments of Foundry Models sold by Azure and Managed Compute deployments. With a name, the tool searches both resource types and can return both when they share that name. Without a name, it combines both lists.	Deployment name (optional)	Deployment details with Kind set to ADM or Managed.
model_deployment_delete	write	Delete either deployment type. The tool checks deployments of Foundry Models sold by Azure first, then Managed Compute deployments.	Deployment name	Deletion confirmation.

Model analytics and recommendations

Compare model benchmarks and get recommendations for switching to more cost-effective or higher-quality models.

Example prompts:

• “Show me benchmark data for all available models.”
• “Compare benchmark performance between GPT-5.4 and GPT-4.”
• “Find models similar to my current GPT-4 deployment.”
• “What models would give me better quality/cost ratio than what I’m using now?”

Tool	Access	Description	Key inputs	Returns
model_benchmark_get	read	Fetch benchmark data for Foundry models.	Model filters (optional)	Benchmark scores, accuracy, cost, and latency metrics.
model_benchmark_subset_get	read	Get benchmark data for specific model name and version pairs.	Model name and version pairs	Benchmark comparison data for specified models.
model_similar_models_get	read	Find similar models based on deployment or model details.	Deployment name or model name	List of similar models with capability comparison.
model_switch_recommendations_get	read	Get model switch recommendations based on benchmark data.	Current deployment name	Recommended models with quality/cost trade-off analysis.

Model monitoring and operations

Track deployment health, monitor metrics, check deprecation status, and view quota usage, including Managed Compute accelerator quota.

Example prompts:

• “Show me the request metrics for my production-chatbot deployment.”
• “Check if any of my deployments are using deprecated model versions.”
• “Show me quota usage across all regions for my subscription.”
• “Show me A100_80GB, H100_80GB, MI300_192GB, and H200_141GB accelerator quota in East US 2.”

Tool	Access	Description	Key inputs	Returns
model_monitoring_metrics_get	read	Get monitoring metrics (requests, latency, errors, quota) for a model deployment.	Deployment name, time range (optional)	Request count, latency percentiles, error rates, and token usage.
model_deprecation_info_get	read	Get deployment info enriched with deprecation and retirement schedules.	Deployment name (optional)	Deployment details with deprecation dates and suggested replacements.
model_quota_list	read	List available deployment quota and usage for a subscription in a region, including Managed Compute accelerator quota.	Subscription, region	Quota limits, current usage, and available capacity per model family or accelerator.

Project connections

Manage connections to external services (Azure OpenAI, Azure Blob Storage, search, and others) within a Foundry project.

Example prompts:

• “List all connections in my Foundry project.”
• “Show me the details for my azure-search connection.”
• “What connection types and authentication methods are supported?”
• “Create a new AzureOpenAI connection called my-openai using AAD auth.”
• “Delete the old-storage connection from my project.”

Tool	Access	Description	Key inputs	Returns
project_connection_list	read	List all connections in a Foundry project, with optional filtering by category or target.	Category filter, target filter (both optional)	List of connections with name, type, and status.
project_connection_get	read	Get a specific connection by name.	Connection name	Connection details including category, target, and auth type.
project_connection_list_metadata	read	List all supported connection categories and authentication types. Call this first to discover valid values.	None	Supported categories (for example, AzureOpenAI, AzureBlobStorage) and auth types (for example, AAD, key).
project_connection_create	write	Create or replace a project connection.	Connection name, category, target, auth type	Created connection details.
project_connection_update	write	Update an existing project connection.	Connection name, fields to update	Updated connection details.
project_connection_delete	write	Delete a project connection by name.	Connection name	Deletion confirmation.

Prompt optimization

Optimize system prompts and developer messages for better LLM performance.

Example prompts:

• “Optimize my system prompt: ‘You are a helpful customer service agent’ using gpt-5.4.”
• “Improve my agent instructions to get more concise responses.”
• “Refine my optimized prompt to also handle follow-up questions.”

Tool	Access	Description	Key inputs	Returns
prompt_optimize	write	Optimize a developer prompt (system message) for better LLM performance using the Azure OpenAI Prompt Optimizer.	Prompt text, target model, refinement instructions (optional)	Optimized prompt text with explanation of changes.

Continuous evaluation

Enable, monitor, and manage continuous evaluation for agents. Continuous evaluation automatically evaluates agent responses on an ongoing basis. The tool auto-detects the agent kind (prompt or hosted) and configures the appropriate evaluation mechanism — evaluation rules for prompt agents, or scheduled evaluation runs for hosted agents.

Example prompts:

• “Enable continuous evaluation for my customer-support-agent using Relevance and Groundedness evaluators.”
• “Set up continuous evaluation for my hosted agent triage-agent to run every 2 hours.”
• “Show me the continuous evaluation configuration for customer-support-agent.”
• “Disable continuous evaluation for my old-test-agent.”
• “Update continuous evaluation for customer-support-agent to sample 50% of responses.”

Tool	Access	Description	Key inputs	Returns
continuous_eval_create	write	Enable or update continuous evaluation for an agent. Auto-detects agent kind and configures the appropriate evaluation mechanism. Creates the evaluation group if needed.	Agent name, evaluator names, model deployment name (for quality evaluators), enabled flag, interval hours (hosted agents), sampling rate (prompt agents), scenario (standard or business)	Continuous evaluation configuration with ID, status, and schedule details.
continuous_eval_get	read	Get continuous evaluation configuration for an agent. Returns all configurations, or filter by scenario for prompt agents.	Agent name, scenario filter (optional)	List of continuous evaluation configurations with evaluator details, schedule, and status.
continuous_eval_delete	write	Delete a continuous evaluation configuration for an agent. Use continuous_eval_get to find configuration IDs.	Configuration ID, agent name	Deletion confirmation.

Example workflows

Model deployment and optimization:

• “Show me the gpt-5.6-sol model in the catalog.”
• “Deploy gpt-5.6-sol as customer-service-bot with 15 capacity units.”
• “Monitor the request latency for my new deployment.”
• “Recommend more cost-effective alternatives based on current usage.”

Managed Compute quota planning:

• “Use model_quota_list to show my Managed Compute accelerator quota and usage for subscription <subscription-id> in East US 2.”
• “Compare the available instances for A100_80GB, H100_80GB, MI300_192GB, and H200_141GB.”
• “I need four instances. Identify which accelerators currently have enough available quota, and remind me that quota availability doesn’t guarantee deployment capacity.”
• “Before any deployment, show me the selected accelerator and instance count for review because Managed Compute allocates billable dedicated GPU capacity.”

Managed Compute deployment lifecycle:

• “Use model_catalog_list to find models advertised for Managed Compute, where isManagedCompute is true.”
• “For openai--gpt-oss-20b, use model_details_get to confirm that isFoundryManagedCompute is true, and show its registry modelAssetId, resolved deployment templates, and accelerators.”
• “Use model_quota_list to check the selected accelerator in East US 2 before deployment.”
• “Review the deployment template, accelerator, and instance count with me. After I confirm the billable dedicated GPU allocation, use model_deploy_managed_compute to create gpt-oss-managed-test with the registry modelAssetId, one resolved deploymentTemplate, and the approved instanceCount.”
• “Use model_deployment_get to monitor gpt-oss-managed-test and verify that its Kind is Managed.”
• “When testing is complete and I confirm cleanup, use model_deployment_delete to delete gpt-oss-managed-test.”

Agent evaluation workflow:

• “List all agents in my project.”
• “Evaluate my customer-support-agent v2 using Relevance, Groundedness, and Safety evaluators.”
• “Compare my baseline evaluation against the new run.”
• “Show me the comparison results with statistical significance.”

Continuous evaluation setup:

• “List all agents in my project.”
• “Enable continuous evaluation for my customer-support-agent using Relevance, Groundedness, and Safety evaluators with the existing gpt-5.6-luna deployment.”
• “Show me the continuous evaluation configuration for customer-support-agent.”
• “Update continuous evaluation to sample 25% of responses with a maximum of 10 runs per hour.”

Resource management and cleanup:

• “List all my current deployments and their usage.”
• “Check which deployments are using deprecated model versions.”
• “Show me my quota usage across all regions.”
• “Delete unused test deployments to free up capacity.”

Preview limitations

Foundry MCP Server is in public preview. The following limitations apply:

• No network isolation — Foundry MCP Server uses the public endpoint https://mcp.ai.azure.com. Resources behind Azure Private Links aren’t accessible. For private MCP connectivity, build your own MCP server and connect it to Agent Service with private networking.
• Data residency — Requests and responses might be processed in EU or US data centers. The server itself doesn’t store data, but cross-region processing can occur.
• No SLA — Preview features don’t include a service-level agreement. Don’t use the server for production workloads that require guaranteed availability.
• Tool set might change — Tools, parameters, and return values might change during the preview period without notice.

For more information, see Supplemental Terms of Use for Microsoft Azure Previews.

Common errors

Error	Cause	Resolution
Access denied	Insufficient Azure RBAC role on the Foundry project or account.	Assign at least Reader for read tools or Contributor for write tools. See RBAC for Microsoft Foundry.
Authentication failure	Expired or invalid Entra ID token.	Sign out and sign back in to your Azure account in Visual Studio Code, or the tool you’re using.
Quota exceeded	Not enough capacity to create a deployment or run an evaluation.	Use model_quota_list to check available quota before the operation.
Resource not found	The specified deployment, dataset, agent, or connection doesn’t exist.	Use the corresponding get or list tool to verify the resource name.
Private endpoint not reachable	Foundry resources use Azure Private Links that the hosted Foundry MCP Server can’t reach.	Remove private endpoint restrictions, use SDKs/REST APIs, or use a custom MCP server with Agent Service private networking.

For more troubleshooting guidance, see Foundry MCP Server security and best practices.

Related content

• Get started with Foundry MCP Server
• Learn how to build your own MCP server
• Review security best practices for MCP servers
• Learn how to submit your MCP server for certification at Microsoft MCP server certification overview.