Can I search for specific coding decisions?

Yes, you can use the /team-shinchan:memories search [keyword] command to quickly find specific entries related to any technical topic or decision.

How does Team-Shinchan acquire new memories?

The system learns automatically after task completion, or you can explicitly teach it new patterns using the /team-shinchan:learn command.

Where are the project memories stored?

All learnings and memories are stored locally in the .shinchan-docs/learnings.md file within your project repository, ensuring data remains under your control.

What should I do if a recorded memory is no longer valid?

If a pattern or preference changes, you can use the /team-shinchan:forget command to remove outdated information and maintain an accurate knowledge base.

What categories of information are tracked?

The skill tracks preferences, coding patterns, project conventions, past mistakes, architectural decisions, and general insights.

Shinchan Memory Manager

Name: Shinchan Memory Manager
Author: seokan-jeong

byseokan-jeong

•

Productivity & Workflow

Organizes and retrieves project-specific learnings, coding patterns, and architectural decisions to maintain consistency across AI agent interactions.

The Shinchan Memory Manager skill acts as a persistent knowledge base for the Team-Shinchan agent ecosystem, allowing users to track preferences, coding patterns, and project conventions over time. By centralizing learnings in a structured format within your repository, it ensures that Claude remains aligned with your project's evolution, helping avoid repeated mistakes and ensuring adherence to established standards. It is particularly valuable for long-term projects where maintaining technical consistency and architectural integrity across multiple sessions is critical.

Key Features

01Filtered views to focus on specific categories like architecture or conventions

02Local persistence via the .shinchan-docs/learnings.md file

03Categorized learning retrieval including preferences, patterns, and mistakes

04Confidence-based prioritization for highly relevant project insights

05Keyword-based search functionality for rapid knowledge access

066 GitHub stars

Use Cases

01Identifying past architectural decisions to maintain consistency in new system designs

02Recalling previous bugs or edge cases documented by the AI to prevent regressions

03Reviewing project-specific coding conventions before starting a new feature to ensure compliance

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add seokan-jeong/team-shinchan memories

For use in Claude.ai and ChatGPT

Download Skill