Context Engineering

You know how AI sessions start sharp and then gradually turn into mush as context fills up? This gives you the actual operations to prevent that: write critical state to files so it survives compaction, select with surgical precision instead of dumping whole modules, compress at task boundaries before you hit 95% and lose coherence, and isolate heavy work to subagents so your main session stays clean. The token budget breakdowns are practical, the compaction triggers make sense, and the anti-patterns section will save you from the grep-then-read-entire-file mistake everyone makes. Basically the missing manual for keeping Claude sharp across long sessions.

skills.sh installs: 171
GitHub stars: 2.9K
last updated: Jun 8, 2026

View on GitHub

Install

$ npx -y skills add rohitg00/pro-workflow --skill context-engineering --agent claude-code

Installs into .claude/skills of the current project.

Files

Context Engineering

Four operations control everything about how context flows through an AI coding session. Master them and you control the quality of every response.

The Four Operations

1. Write — Persist Info Outside Context

Move information out of the context window into durable storage so it survives compaction and session boundaries.

Where to write:

Target	When	Example
CLAUDE.md	Permanent project rules	"Always use pnpm, never npm"
NOTES.md / scratchpad	Working state for current task	Architecture decisions, open questions
`.claude/memory/`	Learnings and patterns	`[LEARN]` rules from corrections
External files	Data too large for context	Test plans, migration checklists

Pattern — Scratchpad workflow:

1. Start complex task → create NOTES.md with goals and constraints
2. After research → write findings to NOTES.md
3. After compaction → NOTES.md survives, context does not
4. Resume → read NOTES.md to recover full state