// SCAN REPORT

Research Assistant

5 tools. 2 can modify or destroy data without limits.

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

Last updated: 3 Jul 2026

2 can modify or destroy data

3 read-only

5 tools total

Community server · catalogue entry verified 03/07/2026

How to control Research Assistant ↓

TOOL MAP

What Research Assistant exposes to your agents

Read (3) Write / Execute (1) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Research Assistant tools

Destructive · High delete_research_topic This tool irreversibly deletes entire research topics and their associated data from persistent storage (ChromaDB vector database). Write · Medium save_research_data The tool creates or modifies records in a vector database, which is a Write operation.

2 of Research Assistant's 5 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Research Assistant

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

Deny destructive operations

{
  "delete_research_topic": {
    "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

{
  "save_research_data": {
    "limits": [
      {
        "counter": "save_research_data_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "get_topic_info": {
    "limits": [
      {
        "counter": "get_topic_info_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 Research Assistant — 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 RESEARCH ASSISTANT →

Instant setup, no code required.

FULL CATALOGUE

All 5 Research Assistant tools

DESTRUCTIVE 1 tools

Destructive delete_research_topic Delete a research topic and all its data.

WRITE 1 tools

Write save_research_data Save research content to vector database for future retrieval.

READ 3 tools

Read get_topic_info Get detailed information about a research topic. Read list_research_topics List all available research topics (vector databases). Read search_research_data Search through saved research data using semantic similarity.

EXPLORE

Related servers

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

FAQ

Questions about Research Assistant

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

Yes. The Research Assistant server exposes 1 destructive tools including delete_research_topic. 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 Research Assistant? +

The Research Assistant server has 1 write tools including save_research_data. 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 Research Assistant.

How many tools does the Research Assistant MCP server expose? +

5 tools across 3 categories: Destructive, Read, Write. 3 are read-only. 2 can modify, create, or delete data.

How do I enforce a policy on Research Assistant? +

Register the Research Assistant 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 Research Assistant tool call.

Deterministic rules across all 5 Research Assistant tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON RESEARCH ASSISTANT →

Instant setup, no code required.

5 Research Assistant tools catalogued and risk-classified — across an index of 43,000+ MCP servers.