GitHub Repository

Context window optimization for AI coding agents. Sandboxes tool output, 98% reduction. 15 platforms

16,299 starsTypeScript

Context Mode – 315 KB of MCP output becomes 5.4 KB in Claude Code

Name: Context Mode – 315 KB of MCP output becomes 5.4 KB in Claude Code
Availability: InStock
Author: mksglu

by mksglu·Feb 25, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●●BangerSolve My ProblemSlick

Context-aware summarizer cuts Playwright dumps from 56KB to under 1KB with zero friction.

Strengths

•Sits transparently between Claude Code and MCP outputs; zero code changes to existing workflows
•SQLite FTS5 with BM25 ranking for search-powered summaries, not naive truncation
•Session longevity jumps 30 min → 3 hours; directly addresses measurable pain for Claude Code power users

Weaknesses

•Locked to Claude Code MCP ecosystem; limited portability if user switches agents or vendors
•Depends on Claude Code plugin stability and marketplace; no fallback if integration breaks

Post Description

Every MCP tool call dumps raw data into Claude Code's 200K context window. A Playwright snapshot costs 56 KB, 20 GitHub issues cost 59 KB. After 30 minutes, 40% of your context is gone.

I built an MCP server that sits between Claude Code and these outputs. It processes them in sandboxes and only returns summaries. 315 KB becomes 5.4 KB.

It supports 10 language runtimes, SQLite FTS5 with BM25 ranking for search, and batch execution. Session time before slowdown goes from ~30 min to ~3 hours.

MIT licensed, single command install:

/plugin marketplace add mksglu/claude-context-mode

/plugin install context-mode@claude-context-mode

Benchmarks and source: https://github.com/mksglu/claude-context-mode

Would love feedback from anyone hitting context limits in Claude Code.

NOTE: Because of the timezone issue, I’m resurfacing the same content here.

I’ve already notified support about this via email, so they’re aware.

Original post: https://news.ycombinator.com/item?id=47148025