Skip to content
Get packs

Context Engineering Collection

Guides an agent through designing, debugging and evaluating context and multi-agent systems

Source Download skill
Repository
muratcankoylan/Agent-Skills-for-Context-Engineering
Stars
18,282
Licence
MIT
Files
7 files

What it does

This is a large knowledge pack for AI agents that explains how to manage what information an AI model sees while it works: how to keep context focused, offload data to files, compress long sessions, design tools, coordinate multiple sub-agents, and set up evaluation and operating loops for autonomous agents. Once installed, when a user asks for help building or debugging an agent system, the agent answers using these structured patterns and points to the relevant sub-guides. It is written guidance rather than a tool that performs actions on its own.

Great for

  • Engineers building or debugging AI agent systems that run long tasks
  • Teams designing multi-agent or orchestrator-worker architectures
  • People who want a structured reference on context management and evaluation

Skip it if

  • Non-technical users wanting a finished app or document (it is guidance, not a product)
  • Anyone expecting the skill to run code or change their files automatically, since the SKILL.md contains only instructions in the files provided
  • Production decisions taken without review, because the advice is general and some topics such as latent KV-cache sharing only apply when the stack and models are compatible

Try asking

  • “Help me design a supervisor and worker architecture for my research agent so workers do not get bloated context”
  • “My agent forgets earlier decisions after a long session; how should I compress its context?”
  • “Create an evaluation rubric for my customer-support agent's tool use”
  • “Review my agent harness for rollback rules and human approval boundaries”
Step-by-step guide: how to use Context Engineering CollectionInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 7 original files, the MIT licence and an attribution note.

    context-engineering-collection.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip context-engineering-collection.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip context-engineering-collection.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Ask for a specific pattern (for example compaction or tool consolidation) so the agent applies the relevant section rather than a broad overview
  • Give the agent your actual architecture or failure logs so it can tailor the advice instead of generic guidance
  • Treat recommendations as a starting point and verify them against your own evaluations before changing production systems

What's inside

  • AGENTS.md
  • CHANGELOG.md
  • CLAUDE.md
  • CONTRIBUTING.md
  • README.md
  • requirements-dev.txt
  • SKILL.md

Redistributed unmodified from muratcankoylan/Agent-Skills-for-Context-Engineering under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.