// SCAN REPORT

Morphik MCP

17 tools. 6 can modify or destroy data without limits.

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

Last updated: 11 Jun 2026

6 can modify or destroy data

11 read-only

17 tools total

Community server · catalogue entry verified 11/06/2026

How to control Morphik MCP ↓

TOOL MAP

What Morphik MCP exposes to your agents

Read (11) Write / Execute (5) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Morphik MCP tools

Destructive · High delete-document This tool irreversibly removes data from the Morphik database system. Write · Medium morphik-filters This tool can set or clear metadata filters that affect how data is retrieved. Write · Low ingest-file-from-base64 This tool creates or adds new data (file ingestion) to the Morphik database without irreversibly destroying existing data, qualifying it as Write. Write · Medium ingest-file-from-path This tool creates or modifies data in the Morphik database by adding files—a reversible write operation.

6 of Morphik MCP's 17 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Morphik MCP

PolicyLayer is an MCP gateway — it sits between your AI agents and Morphik MCP, 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

{
  "morphik-filters": {
    "limits": [
      {
        "counter": "morphik-filters_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "check-ingestion-status": {
    "limits": [
      {
        "counter": "check-ingestion-status_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 Morphik MCP — 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 MORPHIK →

Free to start. No card required.

FULL CATALOGUE

All 17 Morphik MCP tools

DESTRUCTIVE 1 tools

Destructive delete-document Delete a document and all of its derived data.

WRITE 5 tools

Write morphik-filters Show, set, or clear typed metadata filters that automatically apply to retrieval and list commands. Write ingest-file-from-base64 Add a file to Morphik Write ingest-file-from-path Add a file to Morphik Write ingest-files-from-paths Add multiple files to Morphik Write ingest-text Add raw text into Morphik so it becomes searchable and available for retrieval.

READ 11 tools

Read check-ingestion-status Check processing status for a document that is being ingested or processed. Read get-document Lookup metadata for a document by ID. Read get-file-info Get detailed information about a file or directory, including size, creation time, and permissions. Use this t Read get-pages-in-range Retrieve an inclusive page range (max 10 pages) from a document. Ideal for reading context around earlier chun Read list-allowed-directories Get a list of directories the MCP server is allowed to access for file operations. Use this to understand wher Read list-directory List files and subdirectories in a specified directory. Results are marked as either [FILE] or [DIR]. Use this Read list-documents List documents with optional metadata filters. Supports high limits, folder scoping, and returning total count Read retrieve-chunks Retrieve the most relevant text or image pages for a given query. Always call this before answering knowledge- Read retrieve-docs Return complete documents that best match a query when you need broader context than individual chunks. Read search-documents Search for documents by filename or title using full-text search. Use this to discover documents before retrie Read search-files Search for files matching a pattern in a directory and its subdirectories. This is useful for finding files wh

EXPLORE

Related servers

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

FAQ

Questions about Morphik MCP

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

Yes. The Morphik MCP server exposes 1 destructive tools including delete-document. 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 Morphik MCP? +

The Morphik MCP server has 5 write tools including morphik-filters, ingest-file-from-base64, ingest-file-from-path. 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 Morphik MCP.

How many tools does the Morphik MCP server expose? +

17 tools across 3 categories: Destructive, Read, Write. 11 are read-only. 6 can modify, create, or delete data.

How do I enforce a policy on Morphik MCP? +

Register the Morphik 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 Morphik MCP tool call.

Deterministic rules across all 17 Morphik MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON MORPHIK →

Free to start. No card required.

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