learnings-summary

Synthesize closed Connect opportunity artifacts into structured learnings and draft updated PDDs.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Closing an ACE opportunity generates scattered artifacts across GDrive, OCS transcripts, and monitoring reports, making it difficult to extract consistent, actionable insights to improve future cycles and decide if iteration is needed.

Core Features & Use Cases

  • Cross-Artifact Synthesis: Aggregates and analyzes all opportunity artifacts including original PDDs, test results, LLO feedback, monitoring reports, and OCS transcripts.
  • Categorized Learnings: Organizes insights into process, content, technical, and relationship learnings to clearly identify areas for improvement.
  • Iteration PDD Drafting: Automatically generates a new PDD seeded with identified learnings when iteration is warranted, reducing manual effort for the next ACE cycle.
  • Use Case: After closing a Connect opportunity, use this skill to compile all post-run data into a structured learnings document and get a ready-to-use PDD for the next iteration if the team decides to revisit the intervention.

Quick Start

Use the learnings-summary skill to compile all artifacts from the completed Connect opportunity 'opp-123' into a structured learnings document and draft an updated PDD if iteration is warranted.

Frequently Asked Questions about learnings-summary

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

FAQPage Schema
How do I compile post-opportunity learnings from scattered artifacts after closing a Connect opportunity?

To compile post-opportunity learnings, synthesize fragmented artifacts like original PDDs, LLO feedback, and OCS transcripts into a structured document categorizing insights across process, content, technical, and relationship domains.

What is the best way to analyze closed opportunity performance against the original PDD?

Analyzing closed opportunity performance against the original PDD involves cross-referencing test results, monitoring reports, and transcripts to extract actionable insights and determine if a new iteration cycle is warranted.

Can I automatically generate an updated PDD for the next ACE cycle from previous learnings?

Yes, you can automatically generate an updated PDD seeded with identified learnings for the next ACE cycle when iteration is warranted, reducing manual drafting effort and storing the document in Google Drive.

How do I categorize opportunity insights into process, content, technical, and relationship domains?

Categorizing opportunity insights into process, content, technical, and relationship domains requires synthesizing scattered post-run data from GDrive, OCS transcripts, and monitoring reports to clearly identify areas for improvement.

Does the learnings summary workflow track state for downstream ACE integration?

Yes, the learnings summary workflow tracks state in run_state.yaml for downstream ACE workflow integration, ensuring the formal learnings document and optional updated PDD connect properly with subsequent lifecycle steps.