# claude_execute

Execute a task using Claude Code on the remote PC. Spawns Claude CLI with full coding capabilities (read/write files, run commands, git ops). Returns structured JSON output.

Agent View of the PolicyLayer registry record for `claude_execute`. HTML page: https://policylayer.com/tools/0motionguy-claude-bridge-mcp/claude-execute

## Facts

- Tool: `claude_execute`
- Server: Claude Bridge (`0motionguy/claude-bridge-mcp`) — https://policylayer.com/tools/0motionguy-claude-bridge-mcp.md
- Homepage: https://github.com/0motionguy/claude-bridge-mcp
- Risk category: Execute (High risk)
- Registry record: grade C, identity unverified
- Server rate-limited: no
- Parameters: 0
- Recommended policy verdict: Rate-limited

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "claude_execute",
    "arguments": {}
  }
}
```

## Why claude_execute is rated High

This tool executes arbitrary tasks on a remote machine via Claude Code CLI, with capabilities spanning file operations, shell command execution, and git operations. While the tool itself supports write/read operations, the primary risk is the 'run commands' capability, which allows execution of arbitrary code whose effects depend entirely on the arguments provided. This falls into the Execute category.

From the tool's own definition: "Tool description states: 'Spawns Claude CLI with full coding capabilities (read/write files, run commands, git ops)' and 'Execute a task using Claude Code on the remote PC.' The explicit mention of 'run commands' and 'full coding capabilities' on a remote PC…"

## Use case

AI agents invoke claude_execute to trigger actions in Claude Bridge. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.

## Recommended policy (PolicyLayer)

Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Claude Bridge:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "claude_execute": {
      "limits": [
        {
          "counter": "claude_execute_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on Claude Bridge (3)

- `claude_git_status` — Read — https://policylayer.com/tools/0motionguy-claude-bridge-mcp/claude-git-status.md
- `claude_query` — Read — https://policylayer.com/tools/0motionguy-claude-bridge-mcp/claude-query.md
- `claude_read_file` — Read — https://policylayer.com/tools/0motionguy-claude-bridge-mcp/claude-read-file.md

## For agents

This record is a snapshot. Live verdicts and the full registry:

- Check every server in your MCP config at once: `npx -y policylayer stack`
- Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md)
- Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events`

---

Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=0motionguy-claude-bridge-mcp · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/0motionguy-claude-bridge-mcp