Email Personalization Agent: QA checklist

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

Deployment checklist

• ☐ Approved proof-point library built and versioned
• ☐ Messaging-angle config mapped per persona
• ☐ Banned-claims compliance list written
• ☐ n8n workflow imported, marked nodes configured, send-capable credentials confirmed absent
• ☐ Demo-data run matches `demo-data/sample-output.json`
• ☐ One draft generated per persona and manually reviewed for differentiation
• ☐ QA checklist completed by a non-builder
• ☐ Rollback tested

Failure-mode verification

• ☐ Verified: Draft gets auto-sent does not occur (mitigation: n8n workflow uses read/draft-only sequencing-tool API scopes; QA test confirms no send-capable credential is attached)
• ☐ Verified: Fabricated stat in a draft does not occur (mitigation: Banned-claims list is versioned and expanded every time a miss is caught; every miss is logged as a governance incident)
• ☐ Verified: Same generic draft for every persona does not occur (mitigation: QA test generates one draft per persona and asserts they differ meaningfully)
• ☐ Verified: Proof point used out of context does not occur (mitigation: Proof-point metadata requires industry + persona tags before it's eligible for selection)

Governance verification

• ☐ Agent writes only to its agent-owned fields: `personalization_quality_score`, `email_compliance_status`, `email_draft_status` (sequencing-tool metadata, not CRM fields — see Implementation Notes)
• ☐ 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)