skills.rest
dimagi-internal
dimagi-internalOFFICIAL
Explore all skills by this author →
GitHub

pdd-to-learn-app-eval

Grade Nova-built Learn training apps against their Product Design Document.

1|2|Updated Apr 1, 2026
Search Tags:#commcare#learn-app-eval#pdd-validation#nova-build#assessment-grading#flw-training#ace-workflow
One-click install
npx skills add https://github.com/dimagi-internal/ace --skill pdd-to-learn-app-eval
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pdd-to-learn-app-eval
Source: https://github.com/dimagi-internal/ace/tree/main/skills/pdd-to-learn-app-eval
Command: npx skills add https://github.com/dimagi-internal/ace --skill pdd-to-learn-app-eval

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

ACE teams building Nova-generated Learn training apps for Connect opportunities lack a calibrated, reliable way to verify that apps not only match their original Product Design Document (PDD) structural requirements, but also actually train and gate frontline worker (FLW) competence for safe field deployment, preventing the common failure mode of apps that pass all PDD checks but fail to teach or assess required skills.

Core Features & Use Cases

  • Dual-axis grading: Evaluates apps on conformance (matches PDD module count, order, assessment score wiring, topic coverage, and archetype rules) and fitness (trains and gates FLW competence per expert deployability standards, with hard fails for missing enforcement, label-only content, or missing required language translations).
  • Edge case handling: Automatically detects human-in-the-loop pending stub builds with no completed app to grade, emitting an incomplete verdict instead of false failures.
  • Standing rule enforcement: Blocks deployment for apps that violate core build standards like missing "Learn app" in the display name or incorrect post-submit form navigation.
  • Use case: For any ACE opportunity in Phase 3 or later with a completed Nova Learn app build, run this eval to get a weighted verdict YAML with per-dimension scores, auto-surfaced blockers and warnings, and scoring calibrated to expert-built deployable training tools.

Quick Start

Use the pdd-to-learn-app-eval skill to grade the Nova-built Learn app for the current ACE opportunity against its PDD and output a deployability verdict with per-dimension scores and flagged issues.

Frequently Asked Questions about pdd-to-learn-app-eval

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

FAQPage Schema
How do I validate a Nova Learn app against a Product Design Document for frontline worker training?▼

Validate a Nova Learn app against a Product Design Document by grading its structural conformance and frontline worker training fitness. This checks module count, assessment wiring, and competence gating to produce a weighted deployability verdict YAML with per-dimension scores and flagged blockers.

What does PDD validation for CommCare Learn apps check for deployment readiness?▼

PDD validation checks structural conformance to the design document and real-world training fitness. It enforces build standards like display name rules and post-submit navigation, applying hard fails for missing enforcement, label-only content, or missing required language translations.

Can I grade a Nova Learn app build that still has human-in-the-loop pending stubs?▼

You can grade builds with human-in-the-loop pending stubs. The evaluation automatically detects incomplete builds with no completed app to grade and emits an incomplete verdict instead of returning false failures or incorrect scores.

What happens if a Learn training app violates core build standards like navigation or naming?▼

If an app violates core build standards like missing the display name tag or incorrect post-submit form navigation, the evaluation blocks deployment. It auto-surfaces these standing rule violations as blockers in the final verdict YAML.

Does the Learn app evaluation score both PDD conformance and actual training effectiveness?▼

The evaluation applies dual-axis grading to score both PDD conformance and training effectiveness. It assesses topic coverage and archetype rules alongside fitness standards, applying pre and post score caps aligned to expert deployability calibration for frontline worker tools.

Related Skills

Explore More Product & Management Skills→
jjacksonjjackson
1

pdd-to-learn-app

Generate a Learn app JSON/CCZ from a PDD via Nova.

Community
Intermediate
dimagi-internaldimagi-internal
1

training-flw-guide-eval

Grade CommCare field worker training guides against deployed app flows and accessibility constraints.

Official
Advanced
dimagi-internaldimagi-internal
1

app-multimedia-coverage

Generate and integrate display images into CommCare Connect form questions.

Official
Advanced
dimagi-internaldimagi-internal
1

training-faq

Generate categorized Markdown training FAQ documents for Connect opportunities.

Official
Advanced