fetch-context

Fetch task-relevant context from project files and memory caches.

1|Updated Jan 20, 2026
One-click install
npx skills add https://github.com/shakefu/contextium --skill fetch-context
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: fetch-context
Source: https://github.com/shakefu/contextium/tree/main/.claude/skills/fetch-context
Command: npx skills add https://github.com/shakefu/contextium --skill fetch-context

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you load only the information necessary to complete a task, reducing cognitive load and speeding up decision making by avoiding context overload.

Core Features & Use Cases

  • Minimal-by-default context loading: fetches only what is directly relevant to the current objective.
  • Context sources-aware: reads from project goals, task scopes, memory caches, and project documents to assemble a concise context snapshot.
  • Deterministic context delivery: ensures the same contextual slice is loaded for a given task for reproducible AI behavior.
  • Use Case: when kicking off a new feature or debugging session, quickly establish the relevant context before taking action.

Quick Start

Run the context fetcher to load task-specific context into your current session: ./agents/context-fetcher.sh

Frequently Asked Questions about fetch-context

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

FAQPage Schema
How do I load minimal context for an AI coding task without overloading the session?

Deterministic context delivery ensures the same contextual slice is loaded for a given task by reading from specific files like context/purpose.md and .claude/claude.md. This reproducible behavior provides a concise contextual snapshot for consistent AI workflows.

How do I fetch task-specific context from project documents before debugging?

Run the context fetcher script to load task-specific context into your current session. It executes via ./agents/context-fetcher.sh to assemble a concise contextual snapshot from project documents and task scopes.

What context sources are needed to load relevant project scopes for AI workflows?

Loading relevant project scopes requires context sources such as context/purpose.md, .claude/claude.md, .contextium/tasks.json, and the task-specific memory cache. These files provide the deterministic data needed for concise AI context loading.

Does minimal context loading work with existing memory caches and task files?

Minimal context loading works with existing memory caches and task files by reading sources like .contextium/tasks.json. It applies to code, docs, and task scopes to fetch only directly relevant information for the current objective.