skills.rest
ehfazrezwan
ehfazrezwanCOMMUNITY
Explore all skills by this author →
GitHub

remember

Store categorized user preferences and project decisions via the Neuralscape MCP remember tool.

5|1|Updated Feb 10, 2026
Search Tags:#knowledge graph#project context#agent memory#user preferences#persistent storage#fact saving#MCP tool
One-click install
npx skills add https://github.com/ehfazrezwan/neuralscape --skill remember-ehfazrezwan
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: remember
Source: https://github.com/ehfazrezwan/neuralscape/tree/main/neuralscape-plugin/skills/remember
Command: npx skills add https://github.com/ehfazrezwan/neuralscape --skill remember-ehfazrezwan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AI assistants frequently forget important user preferences, project decisions, and technical conventions across sessions, leading to repeated context-gathering questions, inconsistent behavior, and wasted time re-establishing context for recurring tasks.

Core Features & Use Cases

  • MCP-Native Storage: Uses the Neuralscape MCP remember tool for reliable, async memory writes with no need for manual REST calls.
  • Single Fact Storage: Save one categorized, durable fact (preference, decision, convention, etc.) to a persistent memory layer with automatic deduplication to avoid duplicate entries.
  • Context-Aware Scoping: Automatically routes memories to the correct user and project context, with multi-step fallback resolution for project and user IDs when explicit values are not provided.
  • Privacy & Safety Guardrails: Built-in rules to avoid storing sensitive data like API keys and passwords, prevent prompt injection by only storing neutral facts rather than directives, and default sensitive shared facts to private visibility.
  • Use Case: If a user mentions they prefer TypeScript with strict mode enabled for all new frontend projects, use this skill to save that fact so future agent sessions automatically follow the preference without re-asking.

Quick Start

Use the remember skill to save the fact that the user prefers dark mode for all coding interfaces and categorize it as a preference.

Frequently Asked Questions about remember

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

FAQPage Schema
How do I save user preferences for AI agent memory across sessions?▼

Save user preferences across sessions by writing categorized, durable facts to a persistent memory layer. The agent memory automatically scopes context to the correct user and project to retain preferences for future interactions.

Can I store project decisions and technical conventions in persistent agent memory?▼

Yes, you can store project decisions and technical conventions in persistent agent memory. It saves single durable facts with category classification and automatic deduplication to prevent duplicate entries across conversation sessions.

Does agent memory prevent sensitive data exposure and prompt injection?▼

Agent memory includes built-in guardrails against sensitive data exposure and prompt injection. It avoids storing sensitive data like API keys, stores only neutral facts rather than directives, and defaults sensitive shared facts to private visibility.

What is the best way to retain project context when AI assistants forget technical conventions?▼

Retain project context by saving technical conventions to a persistent memory layer with context-aware scoping. This solves the problem of assistants forgetting technical conventions across sessions by routing memories to the correct project context with multi-step fallback resolution.

How do I avoid duplicate entries when saving facts to persistent storage?▼

Avoid duplicate entries in persistent storage by using automatic deduplication during fact saving. The memory layer performs async, deduplicated writes to ensure identical facts are not stored multiple times in the same user and project context.

Related Skills

Explore More Software Engineering Skills→
2lab-ai2lab-ai

learn

Review conversations to extract learnings and durable facts for persistent memory.

Official
Intermediate
ehfazrezwanehfazrezwan
5

neuralscape-memory

Store and retrieve categorized user preferences and project conventions across sessions.

Community
Advanced
yonatangrossyonatangross
217

memory-fabric

Orchestrate knowledge graph memory operations with entity extraction and deduplication.

Community
Advanced
vut08905vut08905

gh-issues

Fetch GitHub issues, spawn parallel fix agents, and open pull requests.

Community
Advanced