// SCAN REPORT

Managed Agent Control

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

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

Last updated: 2 Jul 2026

6 can modify or destroy data

11 read-only

17 tools total

Community server · catalogue entry verified 02/07/2026

How to control Managed Agent Control ↓

TOOL MAP

What Managed Agent Control 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 Managed Agent Control tools

Destructive · High session_delete This tool irreversibly deletes session data including history and sandbox environment. Execute · High session_message This tool initiates or continues agent execution. Execute · High session_start Starting a managed agent session is an Execute action—it launches an external operation (an agent) that will take actions based on instructions and tool approvals. Execute · High session_interrupt This tool actively interferes with a running agent's execution — stopping its current operation and potentially redirecting it with new instructions.

6 of Managed Agent Control'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 Managed Agent Control

PolicyLayer is an MCP gateway — it sits between your AI agents and Managed Agent Control, 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

{
  "session_archive": {
    "limits": [
      {
        "counter": "session_archive_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "agent_get": {
    "limits": [
      {
        "counter": "agent_get_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 Managed Agent Control — 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 MANAGED AGENT CONTROL →

Instant setup, no code required.

FULL CATALOGUE

All 17 Managed Agent Control tools

DESTRUCTIVE 1 tools

Destructive session_delete Permanently delete a session (history + sandbox). Cannot delete a running session.

EXECUTE 4 tools

Execute session_interrupt Interrupt a running agent, optionally redirecting it with a new instruction. Execute session_message Send a user message to the agent — start work, reply, or continue a turn. Execute session_respond session_respond Execute session_start session_start

WRITE 1 tools

Write session_archive Archive a session: stop accepting new events but keep its history. Reversible-ish.

READ 11 tools

Read agent_get agent_get Read agent_list agent_list Read environment_get Get one environment's full configuration by env_ id (packages, networking). Read environment_list environment_list Read memory_store_get Get one memory store's details by memstore_ id (name, description, timestamps). Read memory_store_list memory_store_list Read session_events session_events Read session_get session_get Read session_list List sessions, optionally filtered to one agent_id and/or statuses. Read vault_get Get one vault's details by vlt_ id (display name, metadata, timestamps). Read vault_list vault_list

EXPLORE

Related servers

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

FAQ

Questions about Managed Agent Control

Can an AI agent delete data through the Managed Agent Control MCP server? +

Yes. The Managed Agent Control 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 Managed Agent Control? +

The Managed Agent Control server has 1 write tools including session_archive. 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 Managed Agent Control.

How many tools does the Managed Agent Control MCP server expose? +

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

How do I enforce a policy on Managed Agent Control? +

Register the Managed Agent Control 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 Managed Agent Control tool call.

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

ENFORCE POLICY ON MANAGED AGENT CONTROL →

Instant setup, no code required.

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