๐ŸŽ Get the FREE AI Skills Starter Guide โ€” Subscribe โ†’
BytesAgainBytesAgain
๐Ÿฆ€ ClawHub

YiHui CONTEXT MODE

by @1yihui

context-mode is an MCP server that saves 98% of your context window by sandboxing tool outputs. It routes large file reads, shell outputs, and web fetches th...

Versionv1.0.0
Downloads497
TERMINAL
clawhub install yihui-context-mode

๐Ÿ“– About This Skill


name: context-mode description: context-mode is an MCP server that saves 98% of your context window by sandboxing tool outputs. It routes large file reads, shell outputs, and web fetches through SQLite + FTS5 instead of flooding the context. Also provides session continuity across context compactions via BM25 retrieval. triggers: - ๅˆ†ๆž้กน็›ฎ - ๆœ‰ๅคšๅฐ‘ๆ–‡ไปถ - ๅธฎๆˆ‘็ ”็ฉถ - ๆฃ€ๆŸฅๆ›ดๆ–ฐ - token ่Š‚็œ - ไธŠไธ‹ๆ–‡ๅŽ‹็ผฉ - context overflow - ๅธฎๆˆ‘ๆœ็ดข tags: - token-optimization - context-management - mcp - productivity compatibility: openclaw | claude-code | vscode-copilot | cursor | gemini-cli license: ELv2 source: https://github.com/mksglu/context-mode

Context Mode

An MCP server that solves the context window problem in AI coding agents. It provides: 1. Context Saving โ€” sandbox tools keep raw data out of context window 2. Session Continuity โ€” SQLite + FTS5 for event tracking 3. Think in Code โ€” program analysis instead of reading files 4. Output Compression โ€” terse output format reducing tokens 65-75%

Available Tools

| Tool | When to Use | Token Savings | |------|-------------|---------------| | ctx_batch_execute | Run multiple commands + auto-search results | 90%+ vs raw exec | | ctx_execute | Single script execution (JS/Python/Shell) | 90%+ vs raw exec | | ctx_execute_file | Run code from a file, return only result | high | | ctx_index | Index docs/knowledge into searchable FTS5 | โ€” | | ctx_search | Search indexed content with BM25 | fast recall | | ctx_fetch_and_index | Fetch URL + index into knowledge base | 90%+ vs raw web fetch |

Decision Rules

Use ctx_batch_execute instead of multiple exec/read calls when:

  • Analyzing multiple files at once
  • Counting/grepping across many files
  • Need command output + search results together
  • Use ctx_execute instead of reading files when:

  • User asks "how many lines/funcs/classes in X"
  • Need to compute something, not just read it
  • Use ctx_fetch_and_index instead of web_fetch when:

  • Researching a topic across multiple pages
  • Full raw content won't fit in context
  • Output Format

    Terse. Drop filler, pleasantries, hedging.

    โŒ "So I ran a command to check the files and found that there are..."
    โœ… "Checked. 3 TypeScript files: src/index.ts (142 lines), src/cli.ts (89 lines)."
    

    OpenClaw Integration

    Install via MCP:

    openclaw mcp set context-mode '{"command":"npx","args":["-y","context-mode"]}'
    openclaw gateway restart