Remote MCP for Claude Code

Give Claude Code real codebase context — without filling the context window

Claude Code is great until the repo gets big: it greps and re-reads files, the context window fills up, auto-compact drops what mattered, and every turn burns tokens. Context Engine gives Claude Code one MCP tool — codebase_context — that returns the code that actually answers the question, inside a token budget you set.

Read the FAQ

Why Claude Code struggles in a large codebase

Out of the box, Claude Code finds code by running grep and reading whole files into the context window. On a small project that is fine. On a large codebase or a monorepo it falls apart: a single grep can dump thousands of lines, the same files get re-read on every turn, and the context window fills until auto-compact throws away detail the agent still needed.

The result is the two complaints developers post most often about Claude Code — it gets slow and expensive on big repos, and it loses track of context right when the task gets hard.

grep / glob brute-force search returns noise, not the code that answers the question
whole-file reads and re-reads inflate token usage every single turn
the context window hits 100% and auto-compact silently drops earlier context
more files in the repo means slower, pricier, less accurate sessions

How Context Engine feeds Claude Code the right code

Context Engine builds a hybrid index of your codebase in the cloud. Every question runs three retrieval paths in parallel — dense semantic vectors, sparse keyword matching, and a symbol graph — then reranks and de-duplicates the candidates and returns a file-level context pack. Claude Code sees exactly one MCP tool, codebase_context, and never has to grep the repo itself.

Crucially, the pack is built to a token budget you pass in. Instead of dumping whole files, it returns the relevant spans, skips files the agent has already seen, and stops at the budget — so the context window stays clean and the token bill stays flat even as the codebase grows.

Set it up in one config block

Add Context Engine to Claude Code the same way you add any MCP server — one block in your MCP config with your API key. Sign in with GitHub, copy your key, and Claude Code gains full-codebase context on the next session. There is a free quota on signup, so you can measure the difference before paying anything.

It keeps up with your codebase

The client syncs your repository incrementally, so only changed files are re-uploaded and re-indexed — the index stays fresh without a full rebuild. Parsing covers 12+ languages at the AST level, and a feedback loop continuously tunes retrieval quality. Whether you drive Claude Code, Cursor, Codex, or any MCP client, the same context pack is one tool call away.

Brute-force search vs. a token-budgeted context pack

	Context Engine	Claude Code grep + read
How code is found	Hybrid retrieval: dense + sparse + symbol graph	grep / glob string match
What comes back	Reranked, de-duped, file-level spans	Whole files, raw matches
Token usage	Capped by the budget you set	Grows with every read and re-read
Context window	Stays clean — only relevant spans	Fills up, triggers auto-compact
Large monorepo	Scales — retrieval is indexed	Slower and noisier as it grows

FAQ

Does this reduce Claude Code token usage?

Yes. Instead of reading whole files and re-reading them each turn, Context Engine returns only the relevant spans within a token budget you set, and skips files the agent has already seen. That keeps the context window clean and cuts the tokens spent finding code.

Does it work on a large monorepo?

Yes — that is the case it is built for. Retrieval runs against a cloud index, so it stays fast and precise as the codebase grows, instead of degrading the way brute-force grep does.

Which languages are supported?

Parsing works at the AST level across 12+ languages, so symbols and structure are understood rather than matched as plain text.

Is it free?

There is a free quota on signup with GitHub, so you can try it on your own repo before paying. Paid plans add more quota.

What does Claude Code actually see?

A single MCP tool, codebase_context. Sync, indexing, retrieval, reranking and the token budget are all handled inside the tool — the agent just asks a question and gets a context pack back.

Give Claude Code the context it has been missing

Sign in with GitHub, add one MCP config block, and Claude Code gets a token-budgeted context pack over your whole codebase on the next session.

← Back to home