Git LFS for your LLM context — 20x Context Savings — an MCP server that keeps big documents out of the model and returns byte-exact quotes for a fraction of the tokens.
Keep big documents outside the model — hand it a small searchable handle instead. The model reads a cheap summary to find what it needs, then pulls back the exact original words to quote, using a fraction of the tokens.
An MCP server for Claude Code, Claude Desktop, and any Model Context Protocol client — with a standalone CLI too.
With uv, add SaveContext to Claude Code in one line — no clone, no pip, no virtualenv:
Then open Claude Code, run /mcp to confirm the tools loaded, and point the model at a big document and just ask — it ingests, looks up, and quotes on its own. To pin where the database lives, add -e SAVECONTEXTDB=/abs/path.db before the --.
Using a different MCP client (Claude Desktop, etc.)? Point its server config at the same command: uvx savecontext. Prefer no MCP at all? The CLI does everything standalone.
Most context tools are lossy summarizers — helpful, but they blur the exact numbers and can't tell you when they're guessing. SaveContext does the two things they can't:
- 🎯 Exact, with receipts. Facts come back as the original characters with their source location, never a paraphrase — every span reconstructs byte-for-byte. A summary is for finding; the quote is for trusting. - 🙅 Says "not found" instead of inventing. When the answer isn't in the document, retrieval is flagged weak and the agent abstains — rather than hallucinating a plausible one. - 📉 Flat cost. The payload to answer questions stays ~7,300 tokens whether the document is 15k or 150k. Pasting scales linearly; this doesn't. - 🔁 One call, many questions. lookup answers a batch of quer
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/AlexanderBoger--SaveContext.html)
Keep the signal. Cut the noise. Local CLI + MCP — fold verbose tool output for Claude Code, Codex, Cursor & Desktop.
RayenMlayeh/jevcrawlMCP server and CLI that crawls websites for AI agents: finds the answer page 2x more often than Crawl4AI, 10x cheaper per answer than letting Claude crawl.
jiawei686/jev-ultrafast-mcpHand a whole browser task off in one call: a decision model drives the page server-side, so a flow costs one call, not a turn per click. Ref-based element tables, code-checked assertions, zero-model m
chris-asmussen/wardenOne MCP server that fronts many MCP servers + Skills behind a small tool set and routes to the best one, keeping an agent's context tiny. Migrate your Claude Code MCPs/skills behind it.
alphaparkinc/genpark-merchant-catalog-sku-enrichment-classifier-skillSKU attribute extraction & Google taxonomy categorization (Anthropic Merchant)
alphaparkinc/genpark-multi-item-goal-basket-planner-skillGoal-driven multi-item basket planner within budget ceilings (Anthropic Shopping)
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.