// SCAN REPORT

A2a Msg

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

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

Last updated: 23 Sept 2026

5 can modify or destroy data

5 read-only

10 tools total

Community server · catalogue entry checked 23/09/2026

How to control A2a Msg ↓

TOOL MAP

What A2a Msg exposes to your agents

Read (5) Write / Execute (4) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous A2a Msg tools

Destructive · High receive_attachment Default first choice for Agent automatic attachment receiving. Takes only attachment_id; no path or overwrite input. Saves to downloads/ und Write · Medium save_attachment Use only when the user specifies a destination path; prefer receive_attachment for automatic receiving. Downloads an attachment addressed to Write · Medium mark_read Explicitly marks one or more messages addressed to current member as read. Write · Medium send_file Sends a single local file (max ${CONFIG.maxAttachmentBytes} bytes) as an attachment to another team member. The bridge reads the file, compu

5 of A2a Msg's 10 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control A2a Msg

PolicyLayer is an MCP gateway — it sits between your AI agents and A2a Msg, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "receive_attachment": {
    "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

{
  "mark_read": {
    "limits": [
      {
        "counter": "mark_read_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "get_unread_summary": {
    "limits": [
      {
        "counter": "get_unread_summary_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 A2a Msg — 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 A2A MSG →

Instant setup, no code required.

FULL CATALOGUE

All 10 A2a Msg tools

DESTRUCTIVE 1 tools

Destructive receive_attachment Default first choice for Agent automatic attachment receiving. Takes only attachment_id; no path or overwrite

WRITE 4 tools

Write mark_read Explicitly marks one or more messages addressed to current member as read. Write save_attachment Use only when the user specifies a destination path; prefer receive_attachment for automatic receiving. Downlo Write send_file Sends a single local file (max ${CONFIG.maxAttachmentBytes} bytes) as an attachment to another team member. Th Write send_message Sends a LAN text message to another team member. Delivery is defined as saved into central DB. WARNING: Receiv

READ 5 tools

Read get_unread_summary Returns a counts-only overview of the current member Read getmsg Lists messages addressed to current member with metadata, short summaries, and ascending-ID pagination cursor. Read list_peers Lists active team members in LAN message box. Returns name and display name. Read read_attachment_text Reads a paginated text chunk of a text-like attachment addressed to the current member (default ${CONFIG.defau Read read_message Reads paginated full text of a specific message addressed to current member. A valid final nonempty chunk of a

EXPLORE

Related servers

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

FAQ

Questions about A2a Msg

Can an AI agent delete data through the A2a Msg MCP server? +

Yes. The A2a Msg server exposes 1 destructive tools including receive_attachment. 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 A2a Msg? +

The A2a Msg server has 4 write tools including mark_read, save_attachment, send_file. 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 A2a Msg.

How many tools does the A2a Msg MCP server expose? +

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

How do I enforce a policy on A2a Msg? +

Register the A2a Msg 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 A2a Msg tool call.

Deterministic rules across all 10 A2a Msg tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON A2A MSG →

Instant setup, no code required.

10 A2a Msg tools catalogued and risk-classified — across an index of 46,500+ MCP servers.

// WHERE THIS COMES FROM

These policies come from A2a Msg's registry record.

The record behind this page: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.

Teams ship this data inside their own products. See what a licence covers →