mem0-remember

Stores user-provided facts into mem0 memory with type classification and metadata.

64.5k|7.6k|Updated Jun 20, 2023
One-click install
npx skills add https://github.com/mem0ai/mem0 --skill mem0-remember
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mem0-remember
Source: https://github.com/mem0ai/mem0/tree/main/integrations/mem0-plugin/.opencode-plugin/opencode-skills/mem0-remember
Command: npx skills add https://github.com/mem0ai/mem0 --skill mem0-remember

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Developers often need to persist decisions, preferences, and learnings discovered during a coding session, but manually recording them breaks flow and loses context. This Skill captures explicit user statements and stores them verbatim into mem0 with structured metadata.

Core Features & Use Cases

  • Verbatim Memory Storage: Stores the exact user-provided text via the mem0 add_memory API with infer=False, skipping LLM extraction since the fact is stated explicitly.
  • Automatic Type Classification: Classifies content into types such as decision, anti_pattern, user_preference, convention, task_learning, or environmental based on content signals.
  • Async Write Confirmation: Polls get_event_status with the returned event_id to confirm the write succeeded and reports the resulting memory ID.
  • Use Case: While working in OpenCode, you realize a project convention like "we always use pnpm instead of npm". Invoke the remember command with that text and it is stored as a convention memory tied to your user, project, and branch.

Quick Start

Ask the assistant to remember a fact by saying something like "remember this: we decided to always use pnpm for package management in this project".

Frequently Asked Questions about mem0-remember

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

FAQPage Schema
How do I save a memory in mem0 from OpenCode?

Invoke the remember command with the text you want stored, for example "/mem0-remember we decided to use pnpm". The skill classifies the content type and calls add_memory with your user ID, project ID, and branch metadata.

What memory types does mem0 support for classification?

This skill classifies stored content into decision, anti_pattern, user_preference, convention, task_learning, or environmental types. Classification is based on content signals like "we decided", "I prefer", or "the convention is".

Why does add_memory return an event_id instead of a memory_id?

Mem0 writes are asynchronous, so add_memory returns an event_id immediately. You must call get_event_status with that event_id to confirm the write succeeded and retrieve the final memory ID.

When should I use infer=False when adding memories to mem0?

Use infer=False when the user explicitly states the fact to store, since no LLM extraction is needed. This stores the text verbatim with confidence 1.0, avoiding reinterpretation of the user's exact wording.

What happens if the mem0 write is still pending?

If get_event_status returns PENDING or processing, the skill reports the event ID as a fallback reference instead of a memory ID. The write will complete asynchronously on the mem0 side.