Use Jaxia with App9 Post
Let an assistant draft and explain social work while keeping sensitive actions behind policy and confirmation.
Jaxia receives the current brand, page, selection, timezone, and user role, then performs social operations only through delegated App9 Post MCP tools. Draft creation can execute directly, while scheduling, publishing, approvals, and inbox replies require explicit confirmation unless the brand policy is already satisfied.
What this means in practice
Jaxia receives the current brand, page, selection, timezone, and user role, then performs social operations only through delegated App9 Post MCP tools. Draft creation can execute directly, while scheduling, publishing, approvals, and inbox replies require explicit confirmation unless the brand policy is already satisfied.
Use the documented brand and account boundaries consistently across the scheduler, REST API, SDK, MCP tools, SSE consumers, and webhooks. That makes the same social operation explainable to an operator and reproducible by an integration.
Implementation checklist
Issue a short-lived delegated token limited to the current user, brand, and allowed tools.
Pass only the context needed for the current turn.
Show suggested actions and tool activity before asking for confirmation.
Recheck role, scope, capability, approval, and billing rules when the tool executes.
Keep the manual scheduler fully available when Jaxia is unavailable.
Example request
This example uses a server-side API key and an explicit brand selection. Replace placeholder values and keep credentials out of browser bundles, client logs, and source control.
curl --request POST \
--url 'https://postapi.app9.co/mcp' \
--header 'Authorization: Bearer APP9_POST_API_KEY' \
--header 'X-App9-Post-Brand: BRAND_ID'Operational details
The assistant does not receive provider access tokens or raw App9 Post API keys.
A five-minute delegated token reduces the blast radius of a copied or delayed tool call.
AI generation costs and social delivery costs remain separately attributable.
Common mistakes to avoid
- Do not send a permanent brand API key to an AI conversation.
- Do not treat a suggested action as confirmed publication.
- Do not make scheduler reliability depend on assistant availability.