# agent_run

${DEPRECATED} Delegate to a local coding agent. mode=plan returns a plan; mode=execute starts an execution task.

Agent View of the PolicyLayer registry record for `agent_run`. HTML page: https://policylayer.com/tools/yuga-hashimoto-localant/agent-run

## Facts

- Tool: `agent_run`
- Server: LocalAnt (`yuga-hashimoto/localant`) — https://policylayer.com/tools/yuga-hashimoto-localant.md
- Homepage: https://github.com/yuga-hashimoto/localant
- 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": "agent_run",
    "arguments": {}
  }
}
```

## Why agent_run is rated High

This tool delegates control to a local coding agent and can start execution tasks, making it an Execute category risk. The 'execute' mode explicitly triggers automated code execution. While marked deprecated, it remains capable of running arbitrary tasks with side effects.

From the tool's own definition: "Tool description states 'Delegate to a local coding agent' with 'mode=execute starts an execution task.' This directly triggers external code execution whose effects depend on the agent's behavior and the arguments passed."

## Use case

AI agents invoke agent_run to trigger actions in LocalAnt. 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 LocalAnt:

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

## Other tools on LocalAnt (215)

- `adb_clear_logcat` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/adb-clear-logcat.md
- `adb_uninstall` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/adb-uninstall.md
- `fs_delete_file_with_approval` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/fs-delete-file-with-approval.md
- `git_reset` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/git-reset.md
- `git_reset_hard` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/git-reset-hard.md
- `git_restore_file` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/git-restore-file.md
- `mcp_server_unregister` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/mcp-server-unregister.md
- `secret_remove` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/secret-remove.md
- `shell_remove_allowed_command` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/shell-remove-allowed-command.md
- `skill_uninstall` — Destructive — https://policylayer.com/tools/yuga-hashimoto-localant/skill-uninstall.md
- `adb_input_text` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/adb-input-text.md
- `adb_install_apk` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/adb-install-apk.md
- `adb_keyevent` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/adb-keyevent.md
- `adb_shell` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/adb-shell.md
- `adb_start_app` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/adb-start-app.md
- `adb_stop_app` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/adb-stop-app.md
- `adb_swipe` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/adb-swipe.md
- `adb_tap` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/adb-tap.md
- `approval_approve` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/approval-approve.md
- `autopilot` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/autopilot.md
- `bash` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/bash.md
- `browser_click` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/browser-click.md
- `browser_close` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/browser-close.md
- `browser_evaluate` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/browser-evaluate.md
- `browser_open` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/browser-open.md
- `browser_press` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/browser-press.md
- `browser_scroll` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/browser-scroll.md
- `browser_select` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/browser-select.md
- `browser_type` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/browser-type.md
- `browser_wait_for` — Execute — https://policylayer.com/tools/yuga-hashimoto-localant/browser-wait-for.md
- …and 185 more: https://policylayer.com/tools/yuga-hashimoto-localant.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=yuga-hashimoto-localant · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/yuga-hashimoto-localant