Is it safe to run the indexer multiple times?

Absolutely. The skill uses upsert logic, meaning it updates existing entries and adds new ones without corrupting your database.

How does the indexer identify learning files?

The skill scans for .md files in designated 'learnings' directories and parses them for standard headers like **Topic:** and **Tags:**.

Can I use this across different projects?

Yes, it supports both global learnings (~/.projects/learnings/) and repo-specific learnings, consolidating both into your manifest.

What is a 'gotcha' learning?

A 'gotcha' is a specific type of entry used for pitfalls or non-obvious bugs; these are flagged with a warning emoji in the generated manifest for quick visibility.

Where is the data stored?

Learnings are indexed into a local SQLite database that is automatically initialized on the first run within your environment.

Compound Learning Indexer

Name: Compound Learning Indexer
Author: Connsulting

byConnsulting

•

Aprendizaje y Documentación

Indexes local markdown learning files into a searchable SQLite database and generates a comprehensive manifest of topics and insights.

The Compound Learning Indexer helps developers build a persistent knowledge base by scanning markdown files for metadata like topics, tags, and 'gotchas'. It centralizes fragmented documentation from both global and repository-specific directories into a structured SQLite database, automatically maintaining a MANIFEST.md file that summarizes your technical expertise. This skill is ideal for engineers who want to cultivate a long-term memory of architectural patterns and solutions directly within their development environment, ensuring that hard-won lessons are easily discoverable for future tasks.

Características Principales

01Automated discovery of markdown files in global and repo-specific directories

021 GitHub stars

03Structured indexing into a local SQLite database for high-performance retrieval

04Safe upsert logic to ensure the index stays current without creating duplicate entries

05Dynamic MANIFEST.md generation with categorical summaries and keyword counts

06Metadata extraction for Topics, Tags, and specialized 'gotcha' flags

Casos de Uso

01Tracking project-specific 'gotchas' and edge cases to prevent recurring bugs

02Generating at-a-glance summaries of technical domain expertise across multiple repositories

03Building a personal searchable knowledge base of coding patterns and best practices

Características Principales

01Automated discovery of markdown files in global and repo-specific directories

021 GitHub stars

03Structured indexing into a local SQLite database for high-performance retrieval

04Safe upsert logic to ensure the index stays current without creating duplicate entries

05Dynamic MANIFEST.md generation with categorical summaries and keyword counts

06Metadata extraction for Topics, Tags, and specialized 'gotcha' flags

Casos de Uso

01Tracking project-specific 'gotchas' and edge cases to prevent recurring bugs

02Generating at-a-glance summaries of technical domain expertise across multiple repositories

03Building a personal searchable knowledge base of coding patterns and best practices