Context Compressor

Compresses large context before reasoning to save tokens while keeping the evidence you need to answer correctly. Runs locally via Python scripts, no MCP server required. You get three modes: baseline for generic compression, query_guided when you have a specific question, and evidence_aware when correctness matters and you want an explicit sufficiency check. The skill auto-triggers at 80K tokens and blocks new agent spawns at 150K until you compress. Best for chunky RAG payloads, massive codebases, or long chat histories where you're burning tokens on repeated context. The evidence validation is honest: if compressed output looks too thin, it tells you instead of guessing.

skills.sh installs: 271
GitHub stars: 45
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add oimiragieo/agent-studio --skill context-compressor --agent claude-code

Installs into .claude/skills of the current project.

Files

Context Compressor

Use this skill when the problem is mostly "too much context" rather than "not enough capability."

This skill is a self-contained local package. It does not require the MCP server. Prefer it when you need quick token profiling, local compression, evidence checks, or a reproducible compression workflow inside the repository.

What this skill does

Use the bundled Python scripts to:

measure raw versus compressed token usage
compress large text or adapted JSON payloads
preserve query-relevant evidence instead of doing naive summarization
check whether compressed output still contains enough evidence to answer safely
support cache-friendly prompt assembly by keeping stable context early and volatile query text late