// SCAN REPORT

Iris MCP

17 tools. 11 can modify or destroy data without limits.

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

Last updated: 1 Jul 2026

11 can modify or destroy data

6 read-only

17 tools total

Community server · catalogue entry verified 01/07/2026

How to control Iris MCP ↓

TOOL MAP

What Iris MCP exposes to your agents

Read (6) Write / Execute (10) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Iris MCP tools

Destructive · High session_delete This tool permanently deletes session data, which cannot be undone. Execute · Medium session_cancel Cancelling a running operation is an active intervention that terminates an in-progress process. Execute · High session_reboot This tool triggers execution of a session restart operation whose effects depend on which session is targeted. Execute · High team_launch The tool initiates and manages running processes ('ensuring its process is active').

11 of Iris 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 Iris MCP

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

Deny destructive operations

{
  "session_delete": {
    "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

{
  "quick_message": {
    "limits": [
      {
        "counter": "quick_message_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "get_date": {
    "limits": [
      {
        "counter": "get_date_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 Iris 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 IRIS →

Instant setup, no code required.

FULL CATALOGUE

All 17 Iris MCP tools

DESTRUCTIVE 1 tools

Destructive session_delete Delete a session permanently.

EXECUTE 8 tools

Execute ask_message Ask a question to a team and wait for their response. Execute permissions__approve Permission approval handler for Claude Code Execute session_cancel Cancel a running session operation. Execute session_fork Fork a session into a new terminal window for manual interaction. Execute session_reboot Reboot a session to start fresh with a clean slate. Execute team_launch Launch a team by ensuring its process is active. Execute team_wake Wake up a team by ensuring its process is active in the pool. Execute team_wake_all Wake up all configured teams sequentially. Sounds the air-raid siren and brings all teams online.

WRITE 2 tools

Write quick_message Quickly send a message to a team without waiting (async/fire-and-forget). Write send_message Send a message to a team and wait for response. Use this for communication that requires acknowledgment or whe

READ 6 tools

Read get_date Get the current system date and time. Read get_logs Query in-memory logs from the Iris MCP server. Read list_teams List all configured teams. Read session_report View the conversation history for a session. Read team_sleep Put a team to sleep by removing its process from the pool. Terminates the team process and frees resources. Read team_status Get the status of teams (awake/active or asleep/inactive).

EXPLORE

Related servers

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

FAQ

Questions about Iris MCP

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

Yes. The Iris MCP server exposes 1 destructive tools including session_delete. 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 Iris MCP? +

The Iris MCP server has 2 write tools including quick_message, send_message. 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 Iris MCP.

How many tools does the Iris MCP server expose? +

17 tools across 4 categories: Destructive, Execute, Read, Write. 6 are read-only. 11 can modify, create, or delete data.

How do I enforce a policy on Iris MCP? +

Register the Iris 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 Iris MCP tool call.

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

ENFORCE POLICY ON IRIS →

Instant setup, no code required.

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