Context Mode#

An MCP server that solves the context window problem in AI coding agents. It provides:

1. Context Saving — sandbox tools keep raw data out of context window
2. Session Continuity — SQLite + FTS5 for event tracking
3. Think in Code — program analysis instead of reading files
4. Output Compression — terse output format reducing tokens 65-75%

Available Tools#

ToolWhen to UseToken Savingsctx_batch_executeRun multiple commands + auto-search results90%+ vs raw execctx_executeSingle script execution (JS/Python/Shell)90%+ vs raw execctx_execute_fileRun code from a file, return only resulthighctx_indexIndex docs/knowledge into searchable FTS5—ctx_searchSearch indexed content with BM25fast recallctx_fetch_and_indexFetch URL + index into knowledge base90%+ 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.

text

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

OpenClaw Integration#

Install via MCP:

bash

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