EMAIL_SUPPORT →
EMAIL_SUPPORT →
// SUPPORT

Help, without
the hold music_

Every plan gets a 24-hour first-response SLA. Production incidents are answered in minutes by an engineer, not a bot loop.

0
median first reply, all channels
0
first-response SLA, every plan
0
platform uptime, last 90 days
0
support CSAT, trailing quarter
// ROUTES

Fastest path first

SELF_SERVE
Read the docs
Quickstart, agent guides, policy engine and the full API reference. Most answers live here.
/docs →
EMAIL
Open a ticket
Include your workspace slug and, if relevant, the run ID from the audit log. It halves the round-trips.
support@blackoffice.ai →
INCIDENTS
Something's on fire
Mail with subject prefix [URGENT] pages the on-call engineer directly. Reserved for production impact.
page_on_call →
// COMMON_TICKETS

Asked every week

Open the run in the audit vault: every action links to the policy line that authorized it. You can revert the action from the run view, tighten the policy, and the agent re-learns instantly. If anything looks wrong, forward the run ID to support and we'll trace it with you.
Any lane can be flipped to draft in one click from its settings page, or via blackoffice lane pause <name>. Pending actions queue as proposals until you re-enable autonomous execution. Nothing is lost.
Yes. Full exports (documents, ledger postings, audit vault) are available from settings or the API at any time, in CSV and JSONL. Exports are also part of offboarding: leave, and everything comes with you.
Anything blocking money movement, filings or agreed SLAs: a stuck payment run, a filing that won't submit, an integration outage. Use the [URGENT] subject prefix. It pages a human immediately, around the clock.
Mail security@blackoffice.ai with reproduction steps. Disclosures are acknowledged within 24 hours, and we don't pursue good-faith researchers.

Still stuck?_

Write to us. A human reads every ticket, and the median first reply is 42 minutes.