Read Working Memory

This loads a daily briefing from your Nowledge Mem knowledge base at the start of each session. It pulls in active focus areas, priorities, unresolved flags, and recent changes so you don't lose context between conversations or across different tools like Cursor and Codex. The idea is you run `nmem wm read` once when you start working, then Claude references what's relevant as you go. It's meant to solve the cold start problem where every new chat begins from zero. Works with both local and remote Nowledge Mem setups, though you'll need the nmem CLI installed. Skip it if you want a genuinely fresh start or you're working on something completely isolated.

skills.sh installs: 773
GitHub stars: 204
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add nowledge-co/community --skill read-working-memory --agent claude-code

Installs into .claude/skills of the current project.

Files

Read Working Memory

Start every session with context. Use Context Bundle when you need owner identity, AI Identity, active scope, active rules, and Working Memory together. Use Working Memory alone for the lighter daily briefing.

When to Use

At session start:

Beginning of a new conversation
Returning to a project after a break
When context about recent work would help

During session:

User asks "what am I working on?" or "what's my context?"
User references recent priorities or decisions
Need to understand what's been happening across tools

Skip when:

Already loaded this session
User explicitly wants a fresh start
Working on an isolated, context-independent task

Usage

Prefer Context Bundle for startup or multi-agent sessions:

nmem --json context --source-app generic-agent

Multi-agent orchestrators can set NMEM_AGENT_ID="<agent-slug>" before launching the child agent. Add NMEM_SPACE only when that whole run should override the identity's default space. Use NMEM_HOST_AGENT_ID only for advanced external aliases.

Read Working Memory alone when you only need current priorities: