# run_command

Execute a shell command

Agent View of the PolicyLayer registry record for `run_command`. HTML page: https://policylayer.com/tools/agentics-ai-code-mcp/run-command

## Facts

- Tool: `run_command`
- Server: Code MCP Server (`agentics-ai/code-mcp`) — https://policylayer.com/tools/agentics-ai-code-mcp.md
- Homepage: https://github.com/agentics-ai/code-mcp
- Risk category: Execute (High risk)
- Registry record: grade F, 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": "run_command",
    "arguments": {}
  }
}
```

## Why run_command is rated High

Shell command execution is a critical capability that can trigger external operations, system calls, and code execution with effects entirely dependent on the command arguments. An AI agent given this tool without strict guardrails could execute destructive commands (rm -rf), exfiltrate data, or compromise system security.

From the tool's own definition: "Tool name 'run_command' with description 'Execute a shell command' explicitly indicates execution of arbitrary shell commands."

## Use case

AI agents invoke run_command to trigger actions in Code MCP Server. 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 Code MCP Server:

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

## Other tools on Code MCP Server (72)

- `delete_file` — Destructive — https://policylayer.com/tools/agentics-ai-code-mcp/delete-file.md
- `docker_cleanup` — Destructive — https://policylayer.com/tools/agentics-ai-code-mcp/docker-cleanup.md
- `docker_containers` — Destructive — https://policylayer.com/tools/agentics-ai-code-mcp/docker-containers.md
- `remove_allowed_command` — Destructive — https://policylayer.com/tools/agentics-ai-code-mcp/remove-allowed-command.md
- `rollback_session` — Destructive — https://policylayer.com/tools/agentics-ai-code-mcp/rollback-session.md
- `docker_build` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/docker-build.md
- `docker_compose` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/docker-compose.md
- `docker_images` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/docker-images.md
- `docker_networks` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/docker-networks.md
- `docker_run` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/docker-run.md
- `docker_system` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/docker-system.md
- `docker_volumes` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/docker-volumes.md
- `git_push` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/git-push.md
- `install_dependencies` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/install-dependencies.md
- `npm_command` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/npm-command.md
- `pip_install` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/pip-install.md
- `run_custom_tool` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/run-custom-tool.md
- `run_javascript` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/run-javascript.md
- `run_python` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/run-python.md
- `run_tests` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/run-tests.md
- `secure_run_command` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/secure-run-command.md
- `secure_run_command_sequence` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/secure-run-command-sequence.md
- `start_coding_session` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/start-coding-session.md
- `start_server` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/start-server.md
- `stop_server` — Execute — https://policylayer.com/tools/agentics-ai-code-mcp/stop-server.md
- `end_coding_session` — Other — https://policylayer.com/tools/agentics-ai-code-mcp/end-coding-session.md
- `analyze_code` — Read — https://policylayer.com/tools/agentics-ai-code-mcp/analyze-code.md
- `analyze_file_differences` — Read — https://policylayer.com/tools/agentics-ai-code-mcp/analyze-file-differences.md
- `compare_commits` — Read — https://policylayer.com/tools/agentics-ai-code-mcp/compare-commits.md
- `compare_files` — Read — https://policylayer.com/tools/agentics-ai-code-mcp/compare-files.md
- …and 42 more: https://policylayer.com/tools/agentics-ai-code-mcp.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=agentics-ai-code-mcp · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/agentics-ai-code-mcp