knowledge-synthesis

Abstracts project-specific learnings into reusable cross-project patterns and insights.

1|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/fabioc-aloha/AlexAgent --skill knowledge-synthesis-fabioc-aloha
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: knowledge-synthesis
Source: https://github.com/fabioc-aloha/AlexAgent/tree/main/plugin/skills/knowledge-synthesis
Command: npx skills add https://github.com/fabioc-aloha/AlexAgent --skill knowledge-synthesis-fabioc-aloha

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Valuable lessons learned in one project are often lost or stored too specifically to reuse elsewhere. This Skill provides a structured process to abstract, generalize, connect, and store knowledge at the highest level that remains true, so insights transfer across projects. ## Core Features & Use Cases - Four-step synthesis process: Abstract project-specific details, generalize to the right abstraction level, connect to existing knowledge, and store in the correct format and location. - Insight-to-pattern promotion: A checklist and quality bar for promoting single observations (GI-* insights) into proven patterns (GK-*) validated across two or more projects. - Anti-pattern guardrails: Explicit rules against storing every learning, copy-pasting project specifics, creating near-duplicates, and over-generalizing. - Use Case: After fixing a recurring bug where JSON metadata breaks on file renames, use this Skill to distill the fix into a universal pattern about rename-aware tooling for self-referencing metadata, stored in the global patterns directory. ## Quick Start Use the knowledge-synthesis skill to abstract this session's learnings and promote any reusable insights into global patterns.

Frequently Asked Questions about knowledge-synthesis

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

FAQPage Schema
How do I turn a project lesson into a reusable pattern?▼

Follow the four-step synthesis process: abstract away project-specific details, generalize to the highest level that stays true, connect it to existing patterns, then store it as a GK-* pattern in the patterns directory with a concrete example.

What is the difference between an insight and a pattern?▼

An insight (GI-*) is a single timestamped observation that may not generalize, stored in the insights directory. A pattern (GK-*) is proven in two or more projects, fully abstracted, actionable, and stored in the patterns directory.

When should I promote an insight to a pattern?▼

Promote when the insight has been observed across multiple projects, is abstracted from its original context, has a clear searchable slug, connects to related patterns, includes a concrete example, and the generalization remains true without over-abstraction.

How do I avoid over-generalizing a learning?▼

Test whether the insight would help someone who has never seen the project, and stay specific if generalizing makes it wrong. Include scope and exceptions rather than writing universal 'always do X' rules.

What are common knowledge capture mistakes to avoid?▼

Avoid storing every learning, copy-pasting project specifics, creating near-duplicates of existing patterns, over-generalizing, and skipping concrete examples. Filter for non-obvious insights and extend existing patterns instead of fragmenting knowledge.