> ## Documentation Index
> Fetch the complete documentation index at: https://docs.loongclaw.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Local Model Plus Outbound Delivery

> Use local or self-hosted inference while keeping delivery truthful as outbound-only.

# Local Model Plus Outbound Delivery

Use this playbook when you want local or self-hosted inference but only need
governed outbound delivery instead of a reply-loop service runtime.

This page is intentionally narrow:

* the model runtime stays local or self-hosted
* delivery stays outbound-only
* nothing here is documented as if it joins the runtime-backed gateway owner
  contract

## When This Is The Right Playbook

* local control over the model runtime matters more than hosted convenience
* no inbound or reply-loop channel is required on day one
* the team wants notifications, reports, or alerts delivered outward without
  overclaiming runtime support

## Use A Different Playbook If

| If you actually need...                                | Go here instead                                                                                                                    |
| ------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------- |
| a hosted provider plus a team-chat runtime             | [Volcengine Plus Feishu Or Lark](/use-loong/volcengine-feishu-lark-playbook) or [WeCom Rollout](/use-loong/wecom-rollout-playbook) |
| a dedicated coding lane plus a lightweight bot surface | [BytePlus Coding Plus Telegram](/use-loong/byteplus-coding-telegram-playbook)                                                      |
| provider-only local setup without delivery guidance    | [Provider Guides](/use-loong/provider-guides/index) and [Provider Recipes](/use-loong/provider-recipes)                            |

## Step 1: Bring Up The Local Provider

Ollama example:

```toml theme={null}
active_provider = "local"

[providers.local]
kind = "ollama"
base_url = "http://127.0.0.1:11434"
model = "auto"
```

Health loop:

```bash theme={null}
loong doctor
loong ask --message "Confirm that the local provider is healthy."
```

Operational note:

* if the server effectively serves one production model, pin `model` explicitly
  instead of forcing catalog discovery drama

## Step 2: Add One Outbound Surface

Webhook example:

```toml theme={null}
[webhook]
enabled = true
endpoint_url = { env = "WEBHOOK_ENDPOINT_URL" }
auth_token = { env = "WEBHOOK_AUTH_TOKEN" }
payload_format = "json_text"
payload_text_field = "text"
```

Smoke-test it:

```bash theme={null}
loong channels send webhook --text "local-runtime smoke test"
```

Email example:

```toml theme={null}
[email]
enabled = true
smtp_host = "smtp.example.com"
smtp_username = { env = "EMAIL_SMTP_USERNAME" }
smtp_password = { env = "EMAIL_SMTP_PASSWORD" }
from_address = "loong@example.com"
```

```bash theme={null}
loong channels send email --target "ops@example.com" --text $'Local run is healthy\nEverything is green.'
```

## Step 3: Keep The Boundary Truthful

Rules that matter:

* webhook, email, Slack, Discord, Teams, and similar surfaces are outbound-only
* they do not join the same reply-loop owner model as Feishu / Lark, Telegram,
  Matrix, or WeCom
* do not route this setup description toward `gateway run` as if outbound-only
  delivery became a supervised runtime

## Step 4: Adjust The Outbound HTTP Boundary When Needed

HTTP-backed outbound delivery blocks private or special-use hosts by default.

If you intentionally target a private bridge or loopback service, widen that
boundary explicitly:

```toml theme={null}
[outbound_http]
allow_private_hosts = true
```

Use that only when the private target is intentional.

## Variants

Other local or self-hosted providers fit the same shape:

* `lm_studio`
* `llamacpp`
* `vllm`
* `sglang`

The key point is the same: local inference plus truthful outbound delivery is a
valid public setup shape, but it is not the same thing as a shipped reply-loop
runtime.

## Troubleshooting

| Symptom                                               | What to check                                                                              |
| ----------------------------------------------------- | ------------------------------------------------------------------------------------------ |
| `ask` is unstable before delivery is even involved    | fix the local provider first; do not start with the outbound surface                       |
| a local webhook target is rejected                    | add `[outbound_http] allow_private_hosts = true` only if the private target is intentional |
| the setup starts being described as a channel runtime | correct the docs or runbook wording; this path is outbound-only                            |

## Continue Reading

* Continue to [Common Setups](/use-loong/common-setups) for the setup hub.
* Continue to [Provider Guides](/use-loong/provider-guides/index) for the
  full built-in provider matrix.
* Continue to [Provider Recipes](/use-loong/provider-recipes) for the
  broader local and self-hosted provider recipe set.
* Continue to [Channel Guides](/use-loong/channel-guides/index) for the
  full shipped channel matrix.
* Continue to [Channel Recipes](/use-loong/channel-recipes) for the
  broader outbound-only delivery recipe set.