flux-remember

Route remember requests to CLAUDE.md or .flux/brain/ storage destinations.

7|1|Updated Feb 21, 2026
One-click install
npx skills add https://github.com/Nairon-AI/flux --skill flux-remember
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: flux-remember
Source: https://github.com/Nairon-AI/flux/tree/main/skills/flux-remember
Command: npx skills add https://github.com/Nairon-AI/flux --skill flux-remember

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Routes remember requests to the appropriate storage destination (CLAUDE.md for short, actionable rules or .flux/brain/ for deeper context).

Core Features & Use Cases

  • Route remember instructions to CLAUDE.md for short, per-session rules to enforce lightweight memory.
  • Route deeper contextual memory to .flux/brain/ to persist decisions, rules, and domain knowledge across sessions.
  • Manage session-phase transitions (remember -> idle) during operation and delegate storage to flux-brain when needed.

Quick Start

Route a remember instruction to CLAUDE.md for short-term rules or to the flux-brain for deeper context.

Frequently Asked Questions about flux-remember

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

FAQPage Schema
How do I route session memory rules to CLAUDE.md or persistent storage?

Session memory routing sends short, actionable rules to CLAUDE.md and deeper contextual decisions to .flux/brain/. The system classifies your memory content and delegates the write operation to the appropriate storage destination automatically.

What is the best way to manage session memory across interactive Claude sessions?

Managing session memory across interactive sessions requires routing instructions based on context depth. Short, per-session rules are stored in CLAUDE.md for lightweight enforcement, while complex domain knowledge persists in .flux/brain/ for cross-session access.

When should I use CLAUDE.md versus .flux/brain/ for memory storage?

Use CLAUDE.md for short, actionable rules that need lightweight per-session enforcement. Use .flux/brain/ when you need to persist deeper contextual decisions, rules, and domain knowledge across multiple interactive sessions.

How does memory classification work for Claude session rules?

Memory classification extracts the content from your remember instruction, evaluates the context depth, and categorizes the destination. It determines whether the memory is a simple command for CLAUDE.md or complex context for .flux/brain/, then delegates the write.

Can I automate session-phase transitions for memory management?

Yes, session-phase transitions like moving from remember to idle are handled automatically during operation. The frontmatter defines triggers and phase handling, ensuring consistent memory governance while delegating storage to the flux-brain flow.

Does session memory routing work without additional dependencies?

Session memory routing operates independently without external dependencies. It handles triggers, content extraction, destination classification, and delegates the write operation entirely through the internal flux-brain flow logic.