In this article, we review how the context is saved in context-mode. You will learn:
What is context-mode?
How is context saved?
What is context-mode?
Context mode is an MCP plugin that sandboxes tool output and indexes it into a local FTS5 store. Your AI coding agent searches that store instead of re-sending raw bytes every turn. 17 adapters. 586,000+ developers. Runs entirely on your machine.
I wrote a detailed article about what is Context mode.
How is context saved?
I use deepwiki.com to understand open source codebase. So I opened the to understand open source codebase. So I opened the and asked this question:
How does context-mode save the context?
It gave an answer with references to the actual code snippets and files.
Context-mode achieves the “315 KB → 5.4 KB” saving by never letting raw tool output reach the model’s context window — it executes code in an isolated sandbox subprocess and only returns what the code explicitly prints (console.log, print, etc.), while the full payload either stays inside the sandbox or is persisted to an on-disk FTS5 index for on-demand retrieval.
Well, what’s a FTS5 index? FTS5 (Full-Text Search 5) is a SQLite virtual table module for full-text search. In context-mode, it’s the underlying engine that powers the knowledge base used by ctx_index/ctx_search to store and query indexed content without keeping raw bytes in the model's context. I will write about this in one of upcoming articles.
The concepts provided by Deepwiki are:
Sandboxed execution
Intent-driven filtering for large outputs
Persistent storage instead of raw dump
Response tracking/savings accounting
Concrete numeric evidence
Enforcement layer
Learn more about how context is saved.
About me:
Hey, my name is Ramu Narasinga. Email: ramu.narasinga@gmail.com
I spent 3+ years studying OSS codebases and wrote 400+ articles on what makes the production-grade. Now I'm putting that into practice differently - instead of writing every fix myself, I run coding agents that do it.
How it works? Register your machine as a Runtime, point it at your repo. Agents pick up issues. write the fix, open the PR. You just review, they execute.
Build your coding agents and get more work done in less time at thinkthroo.com
Top comments (0)