// SCAN REPORT

Tensorus MCP

61 tools. 35 can modify or destroy data without limits.

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

Last updated:

35 can modify or destroy data
26 read-only
61 tools total

Community server · catalogue entry verified 01/07/2026

How to control Tensorus MCP ↓

TOOL MAP

What Tensorus MCP exposes to your agents

Read (26) Write / Execute (26) Destructive / Financial (9)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS
Critical Risk

The most dangerous Tensorus MCP tools

Destructive · High delete_computational_metadata This tool irreversibly removes computational metadata associated with a tensor. Destructive · High delete_lineage_metadata This tool performs an irreversible deletion operation on lineage metadata. Destructive · High delete_named_semantic_metadata_for_tensor The tool permanently removes semantic metadata from tensors. Destructive · High delete_quality_metadata The tool permanently deletes metadata associated with tensors.

35 of Tensorus MCP's 61 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Tensorus MCP

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

Deny destructive operations
{
  "delete_computational_metadata": {
    "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
{
  "create_lineage_relationship": {
    "limits": [
      {
        "counter": "create_lineage_relationship_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "aggregate_tensors": {
    "limits": [
      {
        "counter": "aggregate_tensors_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

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

  1. Create a free account and register Tensorus MCP — nothing to install.
  2. Add these rules — paste them, or build them visually. Tune the limits to your setup.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
ENFORCE POLICY ON TENSORUS →

Instant setup, no code required.

FULL CATALOGUE

All 61 Tensorus MCP tools

DESTRUCTIVE 9 tools
Destructive delete_computational_metadata Delete computational metadata for a given tensor descriptor. Destructive delete_lineage_metadata Delete lineage metadata for a given tensor descriptor. Destructive delete_named_semantic_metadata_for_tensor Delete a named piece of semantic metadata for a given tensor descriptor. Destructive delete_quality_metadata Delete quality metadata for a given tensor descriptor. Destructive delete_relational_metadata Delete relational metadata for a given tensor descriptor. Destructive delete_tensor_descriptor Delete a tensor descriptor by its ID. Destructive delete_usage_metadata Delete usage metadata for a given tensor descriptor. Destructive tensorus_delete_dataset Delete an existing dataset. Destructive tensorus_delete_tensor Delete a tensor from a dataset.
EXECUTE 5 tools
Execute execute_nql_query Execute a Natural Query Language query. Execute tensorus_apply_binary_operation Apply a binary TensorOps operation (e.g., add, subtract). Execute tensorus_apply_einsum Apply an einsum operation. Execute tensorus_apply_list_operation Apply a TensorOps list operation such as concatenate or stack. Execute tensorus_apply_unary_operation Apply a unary TensorOps operation (e.g., log, reshape).
WRITE 21 tools
Write create_lineage_relationship Create a lineage relationship between tensors. Write create_semantic_metadata_for_tensor Create semantic metadata for a given tensor descriptor. Write create_tensor_descriptor Create a new tensor descriptor. Write create_tensor_version Create a new version for a given tensor. Write import_tensor_metadata Import tensor metadata with a specified conflict strategy (skip or overwrite). Write patch_computational_metadata Patch computational metadata for a given tensor descriptor. Write patch_lineage_metadata Patch lineage metadata for a given tensor descriptor. Write patch_quality_metadata Patch quality metadata for a given tensor descriptor. Write patch_relational_metadata Patch relational metadata for a given tensor descriptor. Write patch_usage_metadata Patch usage metadata for a given tensor descriptor. Write save_tensor Save a tensor to a dataset. Write tensorus_create_dataset Create a new dataset. Write tensorus_ingest_tensor Ingest a new tensor into a dataset. Write tensorus_update_tensor_metadata Replace metadata for a specific tensor. Write update_named_semantic_metadata_for_tensor Update a named piece of semantic metadata for a given tensor descriptor. Write update_tensor_descriptor Update a tensor descriptor by its ID. Write upsert_computational_metadata Upsert computational metadata for a given tensor descriptor. Write upsert_lineage_metadata Upsert lineage metadata for a given tensor descriptor. Write upsert_quality_metadata Upsert quality metadata for a given tensor descriptor. Write upsert_relational_metadata Upsert relational metadata for a given tensor descriptor. Write upsert_usage_metadata Upsert usage metadata for a given tensor descriptor.
READ 26 tools
Read aggregate_tensors Aggregate tensor metadata based on a grouping field and aggregation function. Read analytics_get_co_occurring_tags Get co-occurring tags based on minimum co-occurrence and limit. Read analytics_get_complex_tensors Get complex tensors based on minimum parent count, transformation steps, and limit. Read analytics_get_stale_tensors Get stale tensors based on a threshold of days and limit. Read backend_connectivity_test Test connectivity to the backend API. Read backend_ping Ping the backend /health endpoint and forward the response. Read connection_test Simple connectivity test returning a static ok status. Read export_tensor_metadata Export tensor metadata for specified tensor IDs or all tensors if IDs are not provided. Read get_all_semantic_metadata_for_tensor Get all semantic metadata for a given tensor descriptor. Read get_child_tensors Get the child tensors for a given tensor in the lineage. Read get_computational_metadata Get computational metadata for a given tensor descriptor. Read get_lineage_metadata Get lineage metadata for a given tensor descriptor. Read get_parent_tensors Get the parent tensors for a given tensor in the lineage. Read get_quality_metadata Get quality metadata for a given tensor descriptor. Read get_relational_metadata Get relational metadata for a given tensor descriptor. Read get_tensor Retrieve a tensor by record ID. Read get_tensor_descriptor Get a tensor descriptor by its ID. Read get_usage_metadata Get usage metadata for a given tensor descriptor. Read list_tensor_descriptors List tensor descriptors with extensive optional filters. Read list_tensor_versions List all versions for a given tensor. Read management_get_metrics Retrieve operational metrics from the Tensorus service. Read management_health_check Perform a health check on the Tensorus service. Read mcp_server_status Check the MCP server's status and current operational mode. Read search_tensors Search for tensors based on a text query, optionally specifying fields to search. Read tensorus_get_tensor_details Retrieve tensor data and metadata. Read tensorus_list_datasets List all available datasets.
EXPLORE

Related servers

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

Yaver 1060 tools
adminautomation
Nodebench 824 tools
adminautomation
UnClick 813 tools
adminautomation
Binance MCP Server 734 tools
adminautomation
Fortimanager 584 tools
adminautomation
Crow 573 tools
adminautomation
Mcp 571 tools
adminautomation
GoHighLevel MCP Server 566 tools
adminautomation
FAQ

Questions about Tensorus MCP

Can an AI agent delete data through the Tensorus MCP server? +

Yes. The Tensorus MCP server exposes 9 destructive tools including delete_computational_metadata, delete_lineage_metadata, delete_named_semantic_metadata_for_tensor. 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 Tensorus MCP? +

The Tensorus MCP server has 21 write tools including create_lineage_relationship, create_semantic_metadata_for_tensor, create_tensor_descriptor. 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 Tensorus MCP.

How many tools does the Tensorus MCP server expose? +

61 tools across 4 categories: Destructive, Execute, Read, Write. 26 are read-only. 35 can modify, create, or delete data.

How do I enforce a policy on Tensorus MCP? +

Register the Tensorus 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 Tensorus MCP tool call.

Deterministic rules across all 61 Tensorus MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON TENSORUS →

Instant setup, no code required.

61 Tensorus MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.