Read cold email replies and bounces by API
email.messages
liveMessages in one of your inboxes, newest first, sent and received alike: a 200-character preview each, with direction and a bounced flag (a hard bounce is a label on your own sent message). Pass message_id to read one message in full. Poll it after a send.
AgentMail:
all inputs ✓
✓ marks a combination proven with a real provider call. Others are wired and unit-tested; their first live run verifies them.
Providers
| Provider | Status |
|---|---|
| AgentMail |
live |
Billing
Unit: per_read · unmatched results billed: no
The exact quote is returned by /v1/inspect before every call and held with headroom on execution. The quote is the price floor: if the provider's actual cost overruns it, the run bills the overrun at the same rate, never above the hold. The unused part of every hold is released.
Run it
curl https://api.jell.run/v1/run \ -H "Authorization: Bearer $JELL_API_KEY" \ -d '{ "capability": "email.messages", "input": { "inbox_id": "..." }, "routing": {"provider": "auto"} }'
Input schema
{
"properties": {
"after": {
"description": "ISO 8601 timestamp with a timezone, e.g. 2026-09-04T13:17:00Z. UTC offsets are accepted and normalized to UTC milliseconds.",
"type": "string"
},
"from": {
"description": "Sender substring filters",
"items": {
"type": "string"
},
"type": "array"
},
"inbox_id": {
"description": "From email.inbox or email.inboxes",
"type": "string"
},
"labels": {
"description": "e.g. [\"received\"] for what came in, [\"sent\"] for your own mail (where a bounce shows as \"bounced\")",
"items": {
"type": "string"
},
"type": "array"
},
"limit": {
"default": 20,
"type": "integer"
},
"message_id": {
"description": "Read this one message in full (text) instead of listing",
"type": "string"
}
},
"required": [
"inbox_id"
],
"type": "object"
}
Normalized output fields
messages · count