How do I change the AI's token budget?

You can run the skill and choose from four presets: Economy (~2000 tokens), Light (~3000), Standard (~4000), or Detailed (~6000) depending on your project complexity.

What does the 'suggest_memories' setting do?

When enabled, Claude will proactively suggest adding new information to your project's memory files when it detects significant changes or new logic.

Where are these memory configuration settings stored?

Settings are saved in a YAML-formatted file at .claude/memory/.memory-config.md within your project directory.

What is the purpose of the memory-config skill?

It allows you to view and modify the settings for the ConKeeper memory plugin, such as how many tokens are allocated for project context and how the AI interacts with your memory files.

ConKeeper Memory Configurator

Name: ConKeeper Memory Configurator
Author: swannysec

byswannysec

生産性とワークフロー

Manages persistent memory settings and token budgets for the ConKeeper AI context plugin.

概要

The memory-config skill provides a streamlined interface for managing how Claude retains and processes project context using the ConKeeper plugin. It allows developers to fine-tune their AI's 'memory' by adjusting token budgets across four distinct presets, toggling automated memory suggestions, and customizing output verbosity. By managing these settings, users can optimize Claude's performance for anything from quick bug fixes in small scripts to deep architectural changes in complex, enterprise-level codebases.

主な機能

Customizable output styles including quiet and explanatory modes
Persistent configuration via .memory-config.md files
Session start auto-load preferences management
Toggle for automated memory addition suggestions
Configurable token budget presets (Economy to Detailed)
0 GitHub stars

ユースケース

Optimizing LLM context window usage for large-scale development projects
Reducing token consumption and API costs during routine maintenance tasks
Standardizing AI memory behavior across a development team's workspace

概要

主な機能

Customizable output styles including quiet and explanatory modes
Persistent configuration via .memory-config.md files
Session start auto-load preferences management
Toggle for automated memory addition suggestions
Configurable token budget presets (Economy to Detailed)
0 GitHub stars

ユースケース

Optimizing LLM context window usage for large-scale development projects
Reducing token consumption and API costs during routine maintenance tasks
Standardizing AI memory behavior across a development team's workspace