// SCAN REPORT

Claude C2

94 tools. 51 can modify or destroy data without limits.

6 destructive tools with no built-in limits. Policy required.

Last updated: 12 Jun 2026

51 can modify or destroy data

43 read-only

94 tools total

Community server · catalogue entry checked 12/06/2026

How to control Claude C2 ↓

TOOL MAP

What Claude C2 exposes to your agents

Read (43) Write / Execute (45) Destructive / Financial (6)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Claude C2 tools

Destructive · Critical persist_remove This tool destroys persistence mechanisms (backdoors, scheduled tasks, registry entries, startup scripts, etc.) that cannot be easily recovered once deleted. Destructive · Critical reg_delete Registry modifications are system-critical and persistent; deleting registry keys can disable security features, break system functionality, or cause data loss. Destructive · Critical rm File and directory deletion is inherently destructive and cannot be undone. Destructive · High timestomp Modifying file timestamps is a well-known anti-forensics/evidence-tampering technique.

51 of Claude C2's 94 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Claude C2

PolicyLayer is an MCP gateway — it sits between your AI agents and Claude C2, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "kill": {
    "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

{
  "cp": {
    "limits": [
      {
        "counter": "cp_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "browser_creds": {
    "limits": [
      {
        "counter": "browser_creds_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 Claude C2 — 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 CLAUDE C2 →

Instant setup, no code required.

FULL CATALOGUE

All 94 Claude C2 tools

DESTRUCTIVE 6 tools

Destructive kill Kill process. Auto-selects client. Destructive mv Move/rename file. Auto-selects client. Destructive persist_remove Remove persistence. Auto-selects client. Destructive reg_delete Delete registry key/value. Auto-selects client. Destructive rm Delete file/directory. Auto-selects client. Destructive timestomp Modify file timestamps. Auto-selects client.

EXECUTE 35 tools

WRITE 10 tools

Write cp Copy file. Auto-selects client. Write mcp_config Read or update MCP server configuration Write mcp_edit_code Edit MCP server code using find/replace Write mcp_restore Restore MCP server from backup Write mcp_write_file Write/overwrite entire file in MCP project Write mkdir Create directory. Auto-selects client. Write reg_write Write registry value. Auto-selects client. Write server_file_write Write file to C2 server (auto-creates parent dirs) Write upload Upload file to client. Auto-selects client. Write write Write to file. Auto-selects client.

READ 43 tools

EXPLORE

Related servers

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

FAQ

Questions about Claude C2

Can an AI agent delete data through the Claude C2 MCP server? +

Yes. The Claude C2 server exposes 6 destructive tools including kill, mv, persist_remove. 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 Claude C2? +

The Claude C2 server has 10 write tools including cp, mcp_config, mcp_edit_code. 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 Claude C2.

How many tools does the Claude C2 MCP server expose? +

94 tools across 4 categories: Destructive, Execute, Read, Write. 43 are read-only. 51 can modify, create, or delete data.

How do I enforce a policy on Claude C2? +

Register the Claude C2 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 Claude C2 tool call.

Deterministic rules across all 94 Claude C2 tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON CLAUDE C2 →

Instant setup, no code required.

94 Claude C2 tools catalogued and risk-classified — across an index of 46,500+ MCP servers.

// WHERE THIS COMES FROM

These policies come from Claude C2'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 →