// SCAN REPORT

Cartridge

20 tools. 7 can modify or destroy data without limits.

7 write tools that can modify data. Rate limits recommended.

Last updated: 30 Jun 2026

7 can modify or destroy data

13 read-only

20 tools total

Community server · catalogue entry verified 30/06/2026

How to control Cartridge ↓

TOOL MAP

What Cartridge exposes to your agents

Read (13) Write / Execute (7) Destructive / Financial (0)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

High Risk

The most dangerous Cartridge tools

Execute · High spreader_run The tool performs an execution action ('Execute one modify-spread-tool iteration') on a target entity. Execute · Medium cartridge_load Loading a cartridge doesn't just retrieve data; it changes the runtime environment by activating new tools, onboarding flows, and personality skins. Write · Medium scene_build The tool creates/configures a scene by combining cartridges, skins, and role assignments. Write · Medium bottle_send This is a Write operation because it creates and sends a message to another vessel, which modifies state in the recipient system.

7 of Cartridge's 20 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Cartridge

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

Rate limit write operations

{
  "bottle_reply": {
    "limits": [
      {
        "counter": "bottle_reply_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "bottle_list": {
    "limits": [
      {
        "counter": "bottle_list_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 Cartridge — 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 CARTRIDGE →

Instant setup, no code required.

FULL CATALOGUE

All 20 Cartridge tools

EXECUTE 2 tools

Execute cartridge_load Load a cartridge by ID, making its tools available Execute spreader_run Execute one modify-spread-tool iteration on a target

WRITE 5 tools

Write bottle_reply Reply to a specific bottle Write bottle_send Send a message-in-a-bottle to another vessel Write scene_build Build a scene with cartridge + skin + role assignments Write scene_export Export current scene as shareable JSON (for git repos) Write skin_apply Apply a personality skin to the active cartridge

READ 13 tools

Read bottle_list List all bottles on a dock Read bottle_read Read bottles addressed to this vessel Read cartridge_list List all available cartridges with metadata Read cartridge_onboard Get onboarding info for a cartridge (tailored for human or agent) Read guardian_check Run health check on a specific vessel or session Read guardian_kill Force-terminate a stuck session Read guardian_log Get watchdog event log Read guardian_status Get fleet health overview — all monitored vessels Read scene_status Get current scene configuration Read skin_list List all available personality skins Read spreader_discover_tiles Discover new tile patterns from iteration logs — repeated sequences become candidates Read spreader_reflect Generate a reflection prompt for the Reasoner vessel to analyze loop patterns Read spreader_status Get current loop statistics — iterations, tokens, success rate

EXPLORE

Related servers

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

FAQ

Questions about Cartridge

How do I prevent bulk modifications through Cartridge? +

The Cartridge server has 5 write tools including bottle_reply, bottle_send, scene_build. 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 Cartridge.

How many tools does the Cartridge MCP server expose? +

20 tools across 3 categories: Execute, Read, Write. 13 are read-only. 7 can modify, create, or delete data.

How do I enforce a policy on Cartridge? +

Register the Cartridge 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 Cartridge tool call.

Deterministic rules across all 20 Cartridge tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON CARTRIDGE →

Instant setup, no code required.

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