[Docs] Correct remaining provider and setup guidance

apps/docs/custom-automations.mdx

@@ -184,4 +184,4 @@ provider-supported thinking budget.
 
 See [Schedule maintenance](/cookbook/scheduled-housekeeping) for copyable
 custom automation prompts, or [Draft product updates](/cookbook/product-updates-newsletter)
-for a custom MCP server and a natural-language schedule.
+for a Resend draft workflow and a natural-language schedule.

apps/docs/how-roomote-works.mdx

@@ -4,20 +4,21 @@ icon: workflow
 description: Understand where work starts, how Roomote runs, and how results come back.
 ---
 
-Roomote is designed to act like a shared teammate inside the workflow your team
-already has. It is not an IDE, a local copilot, or a desktop app. It is a
-reviewable cloud agent that is most at home in chat, where teammates can ask
-questions, assign work, and review updates from the conversations they already
-use.
+Roomote agents answer questions, investigate problems, and carry out reviewable
+work. Use the web dashboard to configure Roomote, start a session, and follow
+its results. You can also connect chat, source control, or Linear to bring
+Roomote into workflows your team already uses.
 
-Roomote also works from your source-control provider, Linear, and the web
-dashboard, but a chat surface such as Slack is the primary day-to-day surface
-for most teams.
+A session holds the conversation and can answer directly using connected tools,
+including supported source-control API reads. When work needs a local checkout,
+code changes, or command execution, Roomote delegates a task to an isolated
+sandbox. The session keeps the conversation and task results together.
 
 ## Shared engineering work
 
-Most natural-language work starts as a Roomote session. The recommended first
-path is to mention Roomote in chat with a scoped ask. Use it for
+Most natural-language work starts as a Roomote session. For your first
+repository task, follow [Run your first task](/first-task) in the web dashboard;
+chat connections are optional. Use Roomote for
 implementation, debugging, investigation, small refactors, test fixes,
 documentation updates, planning, and codebase questions.
 
@@ -70,10 +71,11 @@ session lives.
 | Bitbucket     | Pull request events or `@`-mentions when configured        | Bitbucket activity and session view     |
 | Azure DevOps  | Repository or pull request follow-up when configured       | Azure DevOps activity and session view  |
 
-Chat is the recommended surface for the first session after setup because it
-makes Roomote visible inside an existing team conversation. Use the dashboard
-when you want to continue a session or open delegated task details such as
-logs, diffs, and previews.
+Choose the surface that fits the work. You can start and continue sessions
+entirely in the dashboard, or use a connected chat provider when the request
+belongs in an existing team conversation. Open delegated task details in the
+dashboard to inspect logs, diffs, and previews. See
+[Communications](/communications) when you are ready to connect chat or email.
 
 ## How Roomote stays reviewable
 

apps/docs/integrations/granola.mdx

@@ -25,8 +25,12 @@ do not expire, and cannot read ordinary private notes. They can read:
 - notes visible to everyone in the workspace, including the Team space
 - notes in spaces where **Allow Granola API access** is enabled
 
-Granola enables API access by default for newly created spaces. Review
-**Settings > Spaces** and turn it off anywhere Roomote should not read notes.
+Granola enables API access by default for newly created spaces. Review that
+choice when creating a space. For an existing space, open **Integrations >
+Granola API** in the space header and turn off **Allow access with a workspace
+API key** anywhere Roomote should not read notes. Only workspace admins can
+change this setting. See [Granola's workspace API key
+guide](https://docs.granola.ai/help-center/sharing/integrations/granola-api#workspace-api-keys).
 
 If a workspace key is not available, create a personal key from **Settings >
 Connectors > API keys** and select only **Public notes**, leaving **Personal

apps/docs/integrations/index.mdx

@@ -157,9 +157,9 @@ general API or admin access.
 
 For most teams, this order works well:
 
-1. the required providers:
-   [Comms](/communications), [Source Control](/source-control),
-   [Inference](/models), and [Sandboxes](/compute)
+1. [Inference](/models) for conversations, plus [Source Control](/source-control)
+   and [Sandboxes](/compute) for repository tasks; follow
+   [Run your first task](/first-task) to verify this path in the dashboard
 2. one task-tracking integration, such as Linear, Jira, or Asana, if your team
    already works there
 3. one monitoring or deployment integration, such as Sentry, Grafana, Railway,
@@ -174,3 +174,7 @@ For most teams, this order works well:
 After each connection, run a small task that uses the new context. For example,
 ask Roomote to summarize a linked issue, inspect a monitoring alert, or explain
 which environment a support thread belongs to.
+
+[Communications providers](/communications) are optional. Connect one when you
+want to start work from chat or email, or deliver automation reports there;
+you can use Roomote entirely from the web dashboard.

apps/docs/models.mdx

@@ -331,11 +331,12 @@ list.
 - **Models** controls the provider/model pairs that are available, the default
   model, and specialized model roles.
 
-When you connect an inference provider in setup, Roomote tests the candidate
-credentials and selected model before saving them. Authentication, quota,
-model-access, endpoint, and timeout failures are reported during setup so you
-can correct the connection before starting a task. Providers configured through
-environment variables are validated by the deployment at startup instead.
+API-key connections in setup use the same pre-save credential check described
+above: rejected credentials block saving, while exhausted credit or quota
+produces a warning without blocking setup. Model-access, rate-limit, endpoint,
+and timeout failures do not block that save. A saved connection is not proof
+that its model can run; follow the provider's verification steps and complete
+a small task, including when credentials are managed through environment variables.
 
 For subscription-based providers (ChatGPT, GitHub Copilot, Kimi for Coding,
 xAI Grok / SuperGrok, and Z.AI Coding Plan), the provider row also shows current

apps/docs/providers/communications/agentmail.mdx

@@ -42,9 +42,10 @@ then:
 
 1. **Create the inbox** Roomote should receive mail at (or pick an existing
    one). AgentMail's free tier includes 3 inboxes and 100 emails per day and
-   adds a "Sent via AgentMail" footer to outbound mail; custom domains, with
-   AgentMail-managed SPF/DKIM/DMARC, need a paid plan and are recommended for
-   production.
+   adds a "Sent via AgentMail" footer to outbound mail. The free tier also
+   includes one custom domain with DKIM, SPF, and DMARC support; paid plans
+   remove the footer and provide higher limits. Review [AgentMail's current
+   pricing](https://www.agentmail.to/pricing) before choosing a production plan.
 2. **Create an API key from inside that inbox**, so the key is scoped to it.
    Give it these permissions (or full access): `inbox_read`, `inbox_update`,
    `webhook_read`, `webhook_create`, `webhook_update`, `webhook_delete`,

apps/docs/providers/inference/chatgpt.mdx

@@ -41,5 +41,8 @@ on [ChatGPT pricing](https://chatgpt.com/pricing).
   flow with the same eligible ChatGPT account.
 - **Usage is not displayed.** The unofficial usage endpoint may be unavailable;
   this does not necessarily mean model access is disconnected.
-- **A plan limit is reached.** Wait for the displayed window to reset or connect
-  the OpenAI API provider for separately billed capacity.
+- **A plan limit is reached.** Wait for the displayed window to reset, or connect
+  the [OpenAI API provider](/providers/inference/openai) and disconnect ChatGPT
+  Subscription to use separately billed capacity. Connecting an API key alone
+  does not change the route: the subscription takes precedence for `openai/`
+  models while both are connected.

apps/docs/providers/inference/zai-coding-plan.mdx

@@ -24,7 +24,9 @@ ZAI_CODING_PLAN_API_KEY=...
 ZAI_CODING_PLAN_REGION=global # or china
 ```
 
-Both values are required. Models use the `zai-coding-plan/` prefix, keeping
+The configuration form requires a region selection. For environment-variable
+configuration, `ZAI_CODING_PLAN_REGION` defaults to `global` when unset; set it
+to `china` for a China-region membership key. Models use the `zai-coding-plan/` prefix, keeping
 their membership quota separate from metered `zai/` model IDs.
 
 ## Cost behavior

apps/docs/providers/inference/zai.mdx

@@ -24,8 +24,10 @@ ZAI_API_KEY=...
 ZAI_REGION=global # or china
 ```
 
-Both values are required. Models use the `zai/` prefix; Roomote can also map a
-supported GLM image-capable model to the media role. Check audio and video input
+The configuration form requires a region selection. For environment-variable
+configuration, `ZAI_REGION` defaults to `global` when unset; set it to `china`
+for a BigModel key. Models use the `zai/` prefix; Roomote can also map a
+supported GLM image-capable model to the vision role. Check audio and video input
 support separately if you use those attachments.
 
 ## Cost behavior