How do I install the Nmem Memory Working skill?

Run npx skillmds@latest add nowledge-co/nmem-memory-working in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Nmem Memory Working safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Nmem Memory Working?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Nmem Memory Working free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Nmem Memory Working?

nowledge-co (@nowledge-co) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Nmem Memory Working by nowledge-co — Agent Skill

Name: Nmem Memory Working
Author: nowledge-co

Nmem Memory Working

Read the user's daily Working Memory briefing at session start or when recent priorities matter. This gives Google Antigravity cross-tool continuity without bloating the main prompt. Triggered by /nmem-memory-working.

nowledge-co 79db700 2.4 KB Updated Sep 17, 2026

File contents

Read Working Memory

Always prefer native Model Context Protocol (MCP) or Nowledge FS paths for retrieval to prevent terminal command permission prompts and keep execution silent. Only fall back to CLI commands when MCP tools are disabled.

Preferred Retrieval Hierarchy

Context Bundle (MCP): Call MCP tool read_context_bundle to fetch startup identity, rules, and working memory.
Working Memory (MCP): If context bundle is not needed, call read_working_memory.
Nowledge FS (MCP): Call mem_fs tool with command: "cat", path: "/working-memory/working-memory.md" (or context).
CLI Fallback: Only if MCP is unavailable, execute:
```
nmem --json wm read
```
Or context bundle fallback:
```
nmem --json context --source-app google-antigravity
```

If the runtime already knows the current project or agent lane, add --space "<space name>" to the CLI commands or set space_id in the MCP tools. Multi-agent orchestrators can set NMEM_AGENT_ID="<agent-slug>" before launching. Add NMEM_SPACE only when that whole run should override the identity's default space. Use NMEM_HOST_AGENT_ID only for advanced host-id aliases.

When to Use

At session start
When resuming work after a break
When the user asks what they are focused on now
When the current task clearly depends on recent priorities or active initiatives
When updating focus areas or briefings: Use wm-update entrypoint script.

Updating Working Memory In-Place

To update daily focus areas or briefing in Working Memory safely:

python3 hooks/nmem_entrypoint.py wm-update /path/to/updated_working_memory.md

Or supply raw string content:

python3 hooks/nmem_entrypoint.py wm-update --content "# Focus Areas\n- Priority 1..."

Usage Pattern

Read Context Bundle or Working Memory once near the start of a session.
If Context Bundle was already loaded and includes Working Memory, do not read Working Memory again.
Reuse that context mentally instead of re-reading on every turn.
Refresh or update only if the user asks, active focus areas change, or a long-running session needs it.

---name: nmem-memory-workingdescription: Read the user's daily Working Memory briefing at session start or when recent priorities matter. This gives Google Antigravity cross-tool continuity without bloating the main prompt. Triggered by /nmem-memory-working.---# Read Working MemoryAlways prefer native Model Context Protocol (MCP) or Nowledge FS paths for retrieval to prevent terminal command permission prompts and keep execution silent. Only fall back to CLI commands when MCP tools are disabled.## Preferred Retrieval Hierarchy1. **Context Bundle (MCP)**: Call MCP tool `read_context_bundle` to fetch startup identity, rules, and working memory.2. **Working Memory (MCP)**: If context bundle is not needed, call `read_working_memory`.3. **Nowledge FS (MCP)**: Call `mem_fs` tool with `command: "cat", path: "/working-memory/working-memory.md"` (or `context`).4. **CLI Fallback**: Only if MCP is unavailable, execute: ```bash nmem --json wm read ``` Or context bundle fallback: ```bash nmem --json context --source-app google-antigravity ```If the runtime already knows the current project or agent lane, add `--space "<space name>"` to the CLI commands or set `space_id` in the MCP tools. Multi-agent orchestrators can set `NMEM_AGENT_ID="<agent-slug>"` before launching. Add `NMEM_SPACE` only when that whole run should override the identity's default space. Use `NMEM_HOST_AGENT_ID` only for advanced host-id aliases.## When to Use- At session start- When resuming work after a break- When the user asks what they are focused on now- When the current task clearly depends on recent priorities or active initiatives- **When updating focus areas or briefings**: Use `wm-update` entrypoint script.## Updating Working Memory In-PlaceTo update daily focus areas or briefing in Working Memory safely:```bashpython3 hooks/nmem_entrypoint.py wm-update /path/to/updated_working_memory.md```Or supply raw string content:```bashpython3 hooks/nmem_entrypoint.py wm-update --content "# Focus Areas\n- Priority 1..."```## Usage Pattern- Read Context Bundle or Working Memory once near the start of a session.- If Context Bundle was already loaded and includes Working Memory, do not read Working Memory again.- Reuse that context mentally instead of re-reading on every turn.- Refresh or update only if the user asks, active focus areas change, or a long-running session needs it.

Frequently asked questions