// SCAN REPORT

Azure AI Foundry MCP Server

48 tools. 15 can modify or destroy data without limits.

3 destructive tools with no built-in limits. Policy required.

Last updated: 29 Jun 2026

15 can modify or destroy data

33 read-only

48 tools total

Community server · catalogue entry verified 29/06/2026

How to control Azure AI Foundry MCP Server ↓

TOOL MAP

What Azure AI Foundry MCP Server exposes to your agents

Read (33) Write / Execute (12) Destructive / Financial (3)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Azure AI Foundry MCP Server tools

Destructive · High delete_document Removes a document from the index Destructive · High delete_index Deletes the specified index Destructive · High delete_indexer Deletes the indexer Execute · Medium deploy_model_on_ai_services deploy_model_on_ai_services

15 of Azure AI Foundry MCP Server's 48 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Azure AI Foundry MCP Server

PolicyLayer is an MCP gateway — it sits between your AI agents and Azure AI Foundry MCP Server, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "delete_document": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "add_document": {
    "limits": [
      {
        "counter": "add_document_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "agent_query_and_evaluate": {
    "limits": [
      {
        "counter": "agent_query_and_evaluate_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Azure AI Foundry MCP Server — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON AZURE AI FOUNDRY →

Instant setup, no code required.

FULL CATALOGUE

All 48 Azure AI Foundry MCP Server tools

DESTRUCTIVE 3 tools

Destructive delete_document Removes a document from the index Destructive delete_index Deletes the specified index Destructive delete_indexer Deletes the indexer

EXECUTE 4 tools

Execute deploy_model_on_ai_services deploy_model_on_ai_services Execute execute_dynamic_swagger_action Execute a dynamically generated tool from the Swagger specification. Execute run_agent_eval run_agent_eval Execute run_text_eval run_text_eval

WRITE 8 tools

Write add_document Adds a document to the index Write connect_agent Connect to a specific Azure AI Agent and run a query. Write create_azure_ai_services_account create_azure_ai_services_account Write create_foundry_project create_foundry_project Write create_index Creates an AI Search index Write create_indexer Creates a new indexer Write modify_index Updates an AI Search index with a new index definition Write update_model_deployment update_model_deployment

READ 33 tools

Read agent_query_and_evaluate agent_query_and_evaluate Read fetch_finetuning_status Fetches the status of a fine-tuning job using Azure OpenAI API. Read fk_fetch_local_file_contents Reads the content of a local file and returns it as a string Read fk_fetch_url_contents Fetches the contents of the given HTTP URL Read format_evaluation_report Format evaluation results into a readable report with metrics and Studio URL. Read get_agent_evaluator_requirements Get the required input fields for a specific agent evaluator or all agent evaluators. Read get_data_source Retrieves the details of a specific data source by name Read get_document_count Return the total number of documents in the index Read get_finetuning_job_events MCP-compatible function to retrieve all events for a specific finetuning job. Read get_finetuning_metrics Retrieves fine-tuning metrics if the job has succeeded. Read get_indexer Retrieves the details of a specific indexer by name. Read get_model_details_and_code_samples get_model_details_and_code_samples Read get_model_quotas get_model_quotas Read get_prototyping_instructions_for_github_and_labs get_prototyping_instructions_for_github_and_labs Read get_skill_set Retrieves the details of a specific skill set by name Read get_text_evaluator_requirements Get the required input fields for a specific text evaluator or all text evaluators. Read list_agent_evaluators Returns a list of available agent evaluator names for evaluating agent behaviors. Read list_agents List available agents in the Azure AI Agent Service. Read list_azure_ai_foundry_labs_projects list_azure_ai_foundry_labs_projects Read list_data_sources Retrieves the list of all data source names Read list_deployments_from_azure_ai_services list_deployments_from_azure_ai_services Read list_dynamic_swagger_tools list_dynamic_swagger_tools Read list_finetuning_files Lists all files available for fine-tuning in Azure OpenAI. Read list_finetuning_jobs MCP-compatible function to list all finetuning jobs using Azure OpenAI API. Read list_index_names Retrieves the names of all indexes Read list_index_schemas Retrieves the schemas for all indexes Read list_indexers Retrieves the list of all the names of the indexers Read list_models_from_model_catalog list_models_from_model_catalog Read list_skill_sets Retrieves the list of the names of all skill sets Read list_text_evaluators Returns a list of available text evaluator names for evaluating text outputs. Read query_default_agent Send a query to the default configured Azure AI Agent. Read query_index Search a specific index for documents in that index Read retrieve_index_schema Retrieves the schema for a specific index

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Azure AI Foundry MCP Server

Can an AI agent delete data through the Azure AI Foundry MCP Server MCP server? +

Yes. The Azure AI Foundry MCP Server server exposes 3 destructive tools including delete_document, delete_index, delete_indexer. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Azure AI Foundry MCP Server? +

The Azure AI Foundry MCP Server server has 8 write tools including add_document, connect_agent, create_azure_ai_services_account. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Azure AI Foundry MCP Server.

How many tools does the Azure AI Foundry MCP Server MCP server expose? +

48 tools across 4 categories: Destructive, Execute, Read, Write. 33 are read-only. 15 can modify, create, or delete data.

How do I enforce a policy on Azure AI Foundry MCP Server? +

Register the Azure AI Foundry MCP Server MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Azure AI Foundry MCP Server tool call.

Deterministic rules across all 48 Azure AI Foundry MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON AZURE AI FOUNDRY →

Instant setup, no code required.

48 Azure AI Foundry MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.