replies.preview
Ask how Selda would answer an enquiry, from this workspace’s Brain, without creating a lead or storing a draft. Same writer the real reply uses, so tuning against this tunes the real thing. Stores nothing and sends nothing.
| Endpoint | POST /mcp/run |
| Scope | pipeline |
| Sandbox key | allowed |
| Extra entitlement | none |
| MCP tools | selda_preview_reply |
| Convex function | salesInbox/inboundAutoDraft:previewInboundReply |
A free sandbox key may call it.
Arguments
| Argument | Type | Required | Notes |
|---|---|---|---|
projectId | string | required | |
eventType | string | required | What kind of thing arrived: “quote_requested”, “guide_downloaded”, anything you use. |
payload | any | optional | The form as you would send it, keys and all. Exactly what events.ingest takes. |
who | string | optional | Who it is from, for the greeting. Optional, an anonymous enquiry is a real case. |
company | string | optional | |
source | string | optional | |
language | string | optional | Force a language instead of letting the enquiry decide. For checking one deliberately. |
What Selda fills in
You never send these. The HTTP layer overwrites them from the API key, which is what keeps one organisation’s data out of another’s reach.
| Argument | Filled in from |
|---|---|
orgId | your organisation, resolved from the API key |
Example request
Required arguments only. Values in angle brackets are yours to fill in; the optional ones are in the table above.
curl -X POST https://api.selda.ai/mcp/run \
-H "Authorization: Bearer sk_live_..." \
-H "Content-Type: application/json" \
-d '{ "fn": "replies.preview", "args": { "projectId": "<projectId>", "eventType": "<eventType>" } }'{
"fn": "replies.preview",
"args": {
"projectId": "<projectId>",
"eventType": "<eventType>"
}
}Example response
Every endpoint answers in the same envelope.
{
"value": <the function's return value>,
"request_id": "req_..."
}The shape of value is not documented here. salesInbox/inboundAutoDraft:previewInboundReply declares no returns validator, so there is nothing to derive it from, and a shape written out by hand here is a shape that goes stale without anything noticing. Call it once against a test key and read what comes back.
On failure the body is { "error": { "type", "code", "message", "request_id" }, "request_id" } and request_id is echoed in the X-Request-Id header.
Generated from convex/lib/mcpRegistry.ts. Nothing in this reference sends: launchRun is in no
registry and never will be, so a script can prepare a campaign completely and a person still
presses send in the app.