Context Optimization

muratcankoylan/agent-skills-for-context-engineering

When you're burning tokens on massive tool outputs or hitting context limits mid-task, this skill gives you the playbook: observation masking to replace verbose outputs with compact references, KV-cache optimization to maximize prefix reuse (70%+ hit rates mean real cost savings), compaction at 70% utilization, and context partitioning across sub-agents when a single window won't cut it. The source is thorough on trigger points and tradeoffs. The core insight is solid: most agent contexts are dominated by tool output bloat, and strategic masking plus caching beats throwing more tokens at the problem. Use this when token costs or context constraints are actively blocking you, not as preventive optimization.

skills.sh installs: 449
GitHub stars: 18.3K
last updated: May 26, 2026

View on GitHub

Install

$ npx -y skills add muratcankoylan/agent-skills-for-context-engineering --skill context-optimization --agent claude-code

Installs into .claude/skills of the current project.

Files

Context Optimization Techniques

Context optimization extends the effective capacity of limited context windows through strategic compression, masking, caching, and partitioning. Effective optimization increases useful capacity without requiring larger models or longer windows — but only when applied with measurement discipline. The techniques below are ordered by impact and risk.

When to Activate

Activate this skill when:

Context budgets or token costs constrain task complexity
Observation masking can replace verbose tool outputs with retrievable references
Prefix or KV-cache hit rate needs improvement
Retrieval scoping can reduce irrelevant loaded context
Context partitioning can extend effective capacity across agents
Budget triggers are needed for masking, compaction, or partitioning

SKILL.md on GitHub