mcpField
C
context-loader

context-loader

github.com/mem0aiAgent workflows68.0k116 installsUpdated 2d ago
Repository

About context-loader

context-loader is a read-only skill that searches memories relevant to the current task and presents up to 10 findings as a compact context block. It is intended to prime an agent's context while avoiding repeated information already visible in the session.

Key features

  1. 01Extract topics from the task
  2. 02Search memories once
  3. 03Return up to 10 relevant memories
  4. 04Skip context already in the session
  5. 05Stay silent when no results exist

Use cases

  1. 01Load project decisions before starting work
  2. 02Surface relevant preferences for a task
  3. 03Recall lessons related to a topic

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/mem0ai/mem0 --skill context-loader

After installing, type/context-loaderin Claude Code to confirm it loaded.

Details

From the project README

Context Loader

Pre-fetches relevant memories to prime context before working on a task or topic.

Steps

  1. Extract topics from current message/task. Identify: subject areas, people mentioned, project names, goal references.

  2. Call mem0_memory once with action="search" and a focused question about the task.

  3. Output compact context block (max 10 memories):

context-loader: loaded <N> memories for "<task summary>"
  - [decisions] <content> [mem0:<short_id>]
  - [preferences] <content> [mem0:<short_id>]
  - [lessons] <content> [mem0:<short_id>]
  1. If zero results: output nothing. Don't announce empty context.

Constraints

  • Read-only — never modify or delete memories
  • Max 10 memories returned (most relevant only)
  • Silent on empty — only surfaces findings if relevant context exists
  • Skip memories already visible in current session context

Comments

Markdown supported