managing-context

Identify and load relevant project context documents using keyword and path matching.

1|Updated Feb 19, 2026
One-click install
npx skills add https://github.com/Open330/agt --skill managing-context
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: managing-context
Source: https://github.com/Open330/agt/tree/main/context/context-manager
Command: npx skills add https://github.com/Open330/agt --skill managing-context

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Projects often start tasks without quick access to up-to-date context (plans, architecture, and operational notes). This Skill automatically discovers and loads relevant markdown documentation to inform decisions and reduce cognitive load.

Core Features & Use Cases

  • Auto-detect and load context documents based on task keywords, file paths, and task type.
  • Surface planning, architecture, guides, and operations docs to speed up task initiation.
  • Real-world use case: before implementing a feature, the agent retrieves the latest architecture notes and planning status to align changes.

Quick Start

Instruct the agent to fetch the current project context and present a concise summary of relevant documents.

Frequently Asked Questions about managing-context

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

FAQPage Schema
How do I auto-load project context before starting a new task?

To auto-load project context, this Skill discovers relevant markdown documentation by matching task keywords, file paths, and categories. It surfaces planning, architecture, and operations docs to inform execution and reduce cognitive load.

How does keyword matching select relevant planning and architecture documents?

Keyword matching selects relevant documents by analyzing task text against project markdown files. It uses path alignment, task-type mapping, and recency scoring to identify and load the top context sources for the current work.

What is the best way to surface up-to-date documentation for feature implementation?

The best way to surface up-to-date documentation is instructing the agent to fetch current project context. It retrieves latest architecture notes and planning status by scoring document recency, ensuring changes align with existing plans.

Can I use this to retrieve operations and guides across the whole project?

Yes, you can use this to retrieve operations and guides across the project. It auto-detects relevant context documents based on task type and path alignment, presenting a concise summary of the top matched sources.

Do I need markdown files in my repository to use automatic context loading?

Yes, you need markdown files in your repository for automatic context loading. The Skill identifies and loads relevant markdown documentation containing planning, architecture, and operational notes to inform task execution.