Weekly Revenue Brief Agent: QA checklist

To be completed by a non-builder reviewer before go-live.

Deployment checklist

• ☐ Brief template and section config customized to match leadership's actual meeting format
• ☐ Source connections verified for whichever upstream agents are deployed (#003/#007/#008/#009)
• ☐ n8n workflow imported, publishing destination configured
• ☐ Demo-data run matches `demo-data/sample-output.json`
• ☐ No-invented-metric QA test passes (missing-source section renders as unavailable, not fabricated)
• ☐ Delivery-confirmation check tested
• ☐ QA checklist completed by a non-builder
• ☐ Rollback tested

Failure-mode verification

• ☐ Verified: A metric is fabricated to fill a blank section does not occur (mitigation: Prompt requires a source citation per figure; QA test with intentionally missing source data asserts the section renders as "unavailable," not a number)
• ☐ Verified: Brief goes stale or doesn't publish does not occur (mitigation: n8n workflow includes a delivery-confirmation check; failure alerts the owner)
• ☐ Verified: Questions-for-leadership section is empty every week does not occur (mitigation: QA checklist includes a manual read of the questions section for genuine substance, not filler)
• ☐ Verified: Section attributed to the wrong source does not occur (mitigation: Every populated figure includes its source agent/system inline)

Governance verification

• ☐ Agent writes only to its agent-owned fields: None — this agent reads from CRM and other agents' fields but writes only to the published brief document, not to CRM records.
• ☐ No auto-send, auto-close, auto-route, or auto-create action occurs without the human step described in `agent-spec.md`
• ☐ Confidence/model-version is present on every written record where applicable
• ☐ Demo-data run output matches `demo-data/sample-output.json` in structure

Sign-off

Reviewer name: _______________ Date: _______________ Result: Pass / Fail (attach notes)