Speculo Retro

Extracts, deduplicates, and prioritizes usage friction into GitHub-issue-ready proposals.

2|Updated Jul 6, 2026
One-click install
npx skills add https://github.com/NAMEWTA/learning-open-code --skill speculo-retro-namewta
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Speculo Retro
Source: https://github.com/NAMEWTA/learning-open-code/tree/main/speculo/skills/speculo-retro
Command: npx skills add https://github.com/NAMEWTA/learning-open-code --skill speculo-retro-namewta

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? After using Speculo commands and workflows, friction points (bugs, redundant steps, missing capabilities, doc gaps) are scattered across conversation context, command reports, and change status files, making it hard to turn them into actionable improvements. ## Core Features & Use Cases - Friction Extraction & Classification: Scans conversation context, command reports, change status files, and lessons stores, then classifies each friction point as bug, friction, missing-capability, doc-gap, or ergonomics. - Prioritization & Noise Filtering: Scores items by impact and frequency, merges duplicates by root cause, and drops one-off noise or downgrades usage issues to lessons. - Issue-Ready Proposals: Generates structured proposals with title, type, priority, evidence, root cause, acceptance criteria, and deduplication results against existing GitHub issues. - Use Case: After a week of running Speculo workflows, run a retro to collect all friction encountered, and receive a prioritized list of proposals ready for the calling command to file as GitHub issues. ## Quick Start Ask the AI to run a Speculo retro over this session's command reports and change status files to produce prioritized, deduplicated issue proposals.

Frequently Asked Questions about Speculo Retro

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

FAQPage Schema
How do I turn AI workflow friction into GitHub issues?▼

Run a retro pass that scans conversation context, command reports, and change status files for friction, then classify and prioritize each item. The skill outputs structured proposals with title, evidence, root cause, and acceptance criteria that a calling command files via gh.

What evidence sources does a Speculo retro scan?▼

It scans the current conversation, command report markdown files, change .status.json phase history (revisited, blocked, skipped states), change artifacts like prd.md and slices.md, recorded lessons, and deviations from the persistence contract.

How are friction items prioritized and filtered?▼

Priority equals impact scope times frequency, mapped to critical, high, medium, and low labels. One-off or non-reproducible items are dropped, usage-convention issues are downgraded to lessons, and only actionable medium-or-higher items proceed to drafting.

Does this skill create GitHub issues directly?▼

No. It only returns structured proposals and deduplication results. The calling command handles report writing, user confirmation, and actual gh issue create calls after reviewing the proposals.

How does duplicate detection against existing issues work?▼

After drafting, keywords from each proposal's title and root cause are searched against existing issues via gh issue list. Semantic matches set the disposition to drop or record the duplicate issue number in dup_of.