// SCAN REPORT

OpenLink MCP Server for ODBC

32 tools. 10 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 2 Jul 2026

10 can modify or destroy data

22 read-only

32 tools total

Community server · catalogue entry verified 02/07/2026

How to control OpenLink MCP Server for ODBC ↓

TOOL MAP

What OpenLink MCP Server for ODBC exposes to your agents

Read (22) Write / Execute (9) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous OpenLink MCP Server for ODBC tools

Destructive · Medium clear_cache Clearing a cache irreversibly destroys in-memory data (metadata and query results). Execute · Medium run_readonly_query Although constrained to SELECT/WITH (read-only), the tool actively executes SQL against an ODBC-connected database, making it an Execute action. Execute · High chat_prompt_complete The tool invokes an external OPAL backend service to process and complete a chat prompt. Execute · Critical query_database The tool explicitly executes SQL queries, which places it in the Execute category at minimum.

10 of OpenLink MCP Server for ODBC's 32 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control OpenLink MCP Server for ODBC

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

Deny destructive operations

{
  "clear_cache": {
    "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

{
  "export_readonly_query": {
    "limits": [
      {
        "counter": "export_readonly_query_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "describe_table": {
    "limits": [
      {
        "counter": "describe_table_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 OpenLink MCP Server for ODBC — 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 OPENLINK MCP SERVER FOR ODBC →

Instant setup, no code required.

FULL CATALOGUE

All 32 OpenLink MCP Server for ODBC tools

DESTRUCTIVE 1 tools

Destructive clear_cache Clear in-memory metadata/query cache. Optionally provide a tool name prefix.

EXECUTE 7 tools

Execute chat_prompt_complete Tool to use the OPAL backend to complete chat prompt Execute query_database Execute a SQL query and return results in JSON, JSONL or MD format. Execute query_database_jsonl Execute a SQL query and return results in JSONL format. Execute query_database_md Execute a SQL query and return results in MD format. Execute run_readonly_query Run a read-only SELECT/WITH SQL query against the configured ODBC database. Results are row-limited by default Execute spasql_query Execute a SPASQL query and return results. Execute virtuoso_support_ai Tool to use the Virtuoso AI support function

WRITE 2 tools

Write export_readonly_query Export a read-only SELECT/WITH query result to a local file. Rows are not returned to the LLM context. Write export_table Export a full table/view to a local file. Rows are not returned to the LLM context.

READ 22 tools

Read describe_table Retrieve and return a dictionary containing the definition of a table, including column names, data types, nul Read describe_table_compact Describe a table using compact column metadata designed to reduce LLM context usage. Read filter_table_names Retrieve and return a list containing information about tables whose names contain the substring Read find_relationships Return primary key and foreign key metadata for a table when supported by the ODBC driver. Read get_database_summary Return a compact summary of schemas, table/view counts, sample tables, and sample views. Read get_distinct_values Return distinct values and counts for one column, ordered by frequency descending. Read get_schemas Retrieve and return a list of all schema names from the connected database. Read get_table_count Return COUNT() for a table or view. Read get_table_sample Return rows from a table or view. By default row-limited; use all_rows=true to return all rows. Read get_tables Retrieve and return a list containing information about tables in specified schema, if empty uses connection d Read list_indexes List index/statistics metadata for a table when supported by the ODBC driver. Read list_schemas List available database schemas/catalogs from the configured ODBC connection. Read list_tables List tables from the configured ODBC connection. Optionally filter by schema/catalog. Read list_views List views from the configured ODBC connection. Optionally filter by schema/catalog. Read ping_database Test whether the MCP server can connect to the configured ODBC database. Read search_columns Search for column names containing the given text across tables/views. Useful before writing SQL joins or filt Read search_tables Search for tables or views whose names contain the given text. Read sparql_list_entity_types This query retrieves all entity types in the RDF graph, along with their labels and comments if available. Read sparql_list_entity_types_detailed This query retrieves all entity types in the RDF graph, along with their labels and comments if available. Read sparql_list_entity_types_samples This query retrieves samples of entities for each type in the RDF graph, along with their labels and counts. Read sparql_list_ontologies This query retrieves all ontologies in the RDF graph, along with their labels and comments if available. Read virt_get_schemas Retrieve and return a list of all schema names from the connected Virtuoso database.

EXPLORE

Related servers

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

FAQ

Questions about OpenLink MCP Server for ODBC

Can an AI agent delete data through the OpenLink MCP Server for ODBC MCP server? +

Yes. The OpenLink MCP Server for ODBC server exposes 1 destructive tools including clear_cache. 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 OpenLink MCP Server for ODBC? +

The OpenLink MCP Server for ODBC server has 2 write tools including export_readonly_query, export_table. 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 OpenLink MCP Server for ODBC.

How many tools does the OpenLink MCP Server for ODBC MCP server expose? +

32 tools across 4 categories: Destructive, Execute, Read, Write. 22 are read-only. 10 can modify, create, or delete data.

How do I enforce a policy on OpenLink MCP Server for ODBC? +

Register the OpenLink MCP Server for ODBC 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 OpenLink MCP Server for ODBC tool call.

Deterministic rules across all 32 OpenLink MCP Server for ODBC tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON OPENLINK MCP SERVER FOR ODBC →

Instant setup, no code required.

32 OpenLink MCP Server for ODBC tools catalogued and risk-classified — across an index of 43,000+ MCP servers.