Context Compression

muratcankoylan/agent-skills-for-context-engineering

When your agent session hits hundreds of thousands of tokens and starts forgetting which files it modified, you need compression that doesn't break task continuity. This skill walks through three production approaches: anchored iterative summarization for file-heavy work, opaque compression when you need maximum token savings, and regenerative summaries for readable handoffs. The key insight is optimizing for tokens per task, not per request. If your compression drops a file path and the agent has to re-explore the codebase, you've lost more tokens than you saved. Comes with structured templates, probe-based evaluation tactics, and honest tradeoffs between the approaches.

skills.sh installs: 565
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-compression --agent claude-code

Installs into .claude/skills of the current project.

Files

Context Compression Strategies

When agent sessions generate millions of tokens of conversation history, compression becomes mandatory. The naive approach is aggressive compression to minimize tokens per request. The correct optimization target is tokens per task: total tokens consumed to complete a task, including re-fetching costs when compression loses critical information.

When to Activate

Activate this skill when:

Agent sessions exceed context window limits
Codebases exceed context windows (5M+ token systems)
Designing conversation summarization strategies
Debugging cases where agents "forget" what files they modified
Building evaluation frameworks for compression quality
Creating durable handoff summaries that preserve decisions, files, risks, and next actions

SKILL.md on GitHub