Can I load only specific parts of my context?

Yes, you can 'mount' specific layers (like just 'technology' or just 'company') to keep the active context window focused and efficient.

Does it support languages other than Polish?

According to the SKILL.md, this specific implementation is designed to communicate exclusively in Polish, though the underlying logic manages data structures.

How does it handle outdated information?

The skill uses a 'dynamic' tag with deadlines. When context is loaded, it automatically checks for expired entries and asks the user whether to archive or delete them.

What is the primary purpose of Context Collector?

It provides a framework for Claude to store and retrieve personal and professional information across different sessions, ensuring it always has the necessary context to assist you accurately.

Where is my data stored?

All information is stored locally on your machine in Markdown files located at ~/.claude/context/, keeping your data under your control.

Context Collector

Name: Context Collector
Author: Przemocny

byPrzemocny

•

生産性とワークフロー

Manages and persists multi-layered personal and business context for Claude AI through a systematic 20-category knowledge framework.

Context Collector is a specialized skill designed to provide Claude with a long-term, structured memory. It organizes information into 20 distinct layers—10 private and 10 corporate—stored as Markdown files in a local directory. By utilizing interactive briefing modes, quick-save commands, and intelligent context 'mounting,' it ensures that Claude deeply understands your identity, business strategy, and technical environment. This skill is particularly effective for users who want to avoid repeating background information across different sessions, providing a highly personalized and efficient AI experience.

主な機能

01Granular context mounting to load only the specific knowledge domains needed for a task.

0212 GitHub stars

03Conflict detection and deduplication to ensure data integrity during updates.

0420-layer architecture covering everything from personal values to corporate tech stacks.

05Support for static and dynamic data with automated deadline tracking for temporal entries.

06Interactive briefing mode that uses structured interviews to gather deep context.

ユースケース

01Synchronizing team-wide technical context to ensure the AI uses the correct tech stack and processes.

02Onboarding Claude to a complex business environment including mission, roles, and project history.

03Maintaining a persistent personal knowledge base for consistent coaching and productivity tracking.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add przemocny/generic-skills context-collecter

For use in Claude.ai and ChatGPT

Download Skill