Email Monitor Wrapper

See also: openclaw/bridge/run_wrapper_ainl.py (registry) · docs/ainl_openclaw_unified_integration.md (integration hub)

---

Overview

The email monitor is a lightweight OpenClaw bridge workflow (openclaw/bridge/wrappers/email_monitor.ainl, runner name email-monitor).

By default, the wrapper is disabled in openclaw/bridge/run_wrapper_ainl.py because it depends on the optional openclaw mail plugin. Enable it only if your OpenClaw install has that plugin configured.

When enabled, in production it:

• Fetches unread emails via the email adapter (OpenClaw’s mail check).
• If any unread messages are found, sends a single consolidated Telegram notification showing the count and the subject of the most recent email.
• Runs on a configurable schedule (default: every 15 minutes).

This wrapper replaces the legacy Python email_monitor.py script, demonstrating a simple cron job conversion to AINL-native graph execution.

How it works

1. Schedule: The wrapper declares S core cron "*/15 * * * *" inside the source. Production timing is whatever you configure in OpenClaw (openclaw cron add); keep it aligned with the source.

2. Invocation:

   python3 openclaw/bridge/run_wrapper_ainl.py email-monitor [--dry-run]
   

3. Workflow:

   • Calls R email G to get unread messages (returns JSON with a messages array).
   • Extracts the count and the first message’s subject.
   • Builds a notification: 📧 You have X unread email(s).\nLatest: <subject>.
   • Sends via R queue Put ["notify", message] (OpenClaw Telegram delivery).

   In dry-run mode, queue Put is skipped, but the rest executes.

Installation (from scratch)

1. Ensure the wrapper is present in AI_Native_Lang/openclaw/bridge/wrappers/email_monitor.ainl (bundled).

2. Enable and register the wrapper in openclaw/bridge/run_wrapper_ainl.py by adding or uncommenting the following entry in WRAPPERS:

   "email-monitor": _BRIDGE_DIR / "wrappers" / "email_monitor.ainl",
   

3. Add a cron job (or replace existing Python-based email monitor):

   openclaw cron add \
     --name "Email Monitor — Immediate alerts" \
     --cron "*/15 * * * *" \
     --session-key "agent:default:ainl-advocate" \
     --message "Run: cd /path/to/AI_Native_Lang && python3 openclaw/bridge/run_wrapper_ainl.py email-monitor" \
     --description "AINL email monitor wrapper"
   

   Adjust --session-key to match your agent’s session if different.

4. Test:

   python3 openclaw/bridge/run_wrapper_ainl.py email-monitor --dry-run
   

   The dry-run will print the JSON output; inspect the out value for the chosen branch.

Environment variables

| Variable | Purpose | |----------|---------| | OPENCLAW_TARGET or OPENCLAW_NOTIFY_CHANNEL | Destination for Telegram notifications (chat ID or channel). Required for live delivery. | | AINL_DRY_RUN | Set to 1 to skip side effects; also pass --dry-run. |

Example output

Live Telegram notification:

📧 You have 3 unread email(s).
Latest: Re: Project update – please review

Troubleshooting

| Issue | What to do | |-------|------------| | No Telegram | Confirm OPENCLAW_TARGET or OPENCLAW_NOTIFY_CHANNEL is set and that the OpenClaw message plugin is configured for Telegram. | | Adapter errors | Ensure openclaw mail check works from the command line. The email adapter mirrors openclaw mail check --unread --json. | | Cron not running | Verify the cron job status with openclaw cron list and openclaw cron runs <id>. | | Dry-run shows success but live fails | Check agent session key and permissions; ensure the bridge runner has access to email and queue adapters (registry allowlist). |

Development notes

• The wrapper uses only core AINL ops and the email/queue adapters, making it a minimal conversion example.
• To extend (e.g., add importance filtering, digest mode), edit email_monitor.ainl. The source is heavily commented.
• This pattern can be replicated for other simple poll-and-notify cron jobs.