training-onboarding-email-eval

Grade Phase 6 LLO onboarding email drafts against a weighted 5-dimension rubric.

1|2|Updated Apr 1, 2026
One-click install
npx skills add https://github.com/dimagi-internal/ace --skill training-onboarding-email-eval
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: training-onboarding-email-eval
Source: https://github.com/dimagi-internal/ace/tree/main/skills/training-onboarding-email-eval
Command: npx skills add https://github.com/dimagi-internal/ace --skill training-onboarding-email-eval

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Poorly crafted LLO onboarding emails lead to slow Phase 9 response rates, and producers often miss critical flaws in their own drafts during self-review. This Skill provides an independent, rubric-based evaluation to catch these issues before the email is sent to the Local Lead Operator.

Core Features & Use Cases

  • Weighted 5-Dimension Rubric: Scores emails on warmth, clarity, call-to-action effectiveness, context fidelity, and length discipline, with hard-deduct rules for critical failures like missing CTAs or incorrect org names.
  • Structured Verdict Output: Generates a standardized YAML verdict aligned with ACE's eval template, including severity flags for issues that block email deployment.
  • Calibration Logic: Includes targets to align grading accuracy with real Phase 9 LLO onboarding response rates over time, with plans to recalibrate based on real-world performance data.
  • Use Case: When a Phase 6 producer finishes drafting the LLO onboarding email, run this eval to identify generic tone, buried CTAs, or context drift from the original PDD before the email is sent.

Quick Start

Use the training-onboarding-email-eval skill to grade the latest Phase 6 LLO onboarding email draft and receive a structured verdict with actionable improvement feedback.

Frequently Asked Questions about training-onboarding-email-eval

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I evaluate onboarding emails for engagement and clarity before sending?

Evaluate onboarding emails by grading drafts against a weighted 5-dimension rubric covering warmth, clarity, call-to-action effectiveness, context fidelity, and length discipline to identify flaws before deployment. This independent review catches issues producers often miss during self-review.

What is a rubric-based email evaluation and how does it work?

Rubric-based email evaluation scores drafts across five weighted dimensions, applying hard-deduct rules for critical failures like missing calls-to-action or incorrect organization names. It generates a standardized YAML verdict with severity flags to block poor-quality emails from being sent.

How do I grade LLO onboarding email drafts in ACE Connect workflows?

Grade LLO onboarding email drafts in ACE Connect by running an automated evaluation during Phase 6. This produces a structured verdict with actionable feedback, addressing issues before the first automated contact with the Local Lead Operator prior to Phase 9 onboarding.

Why does onboarding email quality affect Phase 9 response rates?

Onboarding email quality directly impacts Phase 9 response rates because poorly crafted outreach with generic tone, buried calls-to-action, or context drift causes slow LLO replies. Independent rubric grading mitigates this by catching critical flaws before the email is sent.

Can I use automated email grading for ACE Connect outreach workflows?

Yes, you can use automated email grading for ACE Connect outreach workflows by applying a calibration logic designed to align grading accuracy with real-world Phase 9 LLO onboarding response rates over time, ensuring accurate evaluation results.

What are the limitations of automated onboarding email evaluation?

A limitation of automated onboarding email evaluation is that its calibration logic currently relies on targets to align with real-world performance, meaning it requires future recalibration based on actual Phase 9 LLO response data to maintain grading accuracy over time.