开源MCP工具：Context Cooler

Requirements

• Node.js 18+ (for the MCP server)
• Python 3.8+ (for the installer and helper scripts — stdlib only, no pip dependencies)
• SQLite (bundled with Python and Node.js via better-sqlite3)

---

Installation & Updates

Installer Options

python3 install.py                              # Interactive — asks which agents to register
python3 install.py --platform=claude-code       # Register one platform, skip prompt
python3 install.py --platform=all               # Register every supported agent
python3 install.py --non-interactive            # Use defaults, no prompts (for CI)
python3 install.py --dry-run                    # Preview changes without writing
python3 install.py --verify                     # Check installation status
python3 install.py --uninstall                  # Show uninstall notes
python3 install.py --update                     # git pull + rebuild + re-register
python3 install.py --accept-disclaimer          # Skip disclaimer prompt (CI/scripts)
python3 install.py --data-dir /custom/path      # Custom data directory

What the Installer Does

Every install runs these four universal steps:

1. Builds the MCP server (npm install + npx tsc).
2. Registers context-cooler with each selected platform adapter (Claude Code, Cursor, Codex, Gemini, OpenCode, Pretzel Porter, Grok CLI). Each adapter writes atomically (tmp file + rename) to that platform's MCP config file.
3. Initialises SQLite databases (stats.db + sessions.db) under the data directory (default: your home directory, override with $CONTEXT_COOLER_HOME or --data-dir). The directory is auto-created.
4. Records the install timestamp in <data-dir>/context/last-upgrade.txt so ctx_doctor can remind you to upgrade later.

Canonical example

"Across these 47 TypeScript files, find every await that's missing a try/catch."

The 195× reduction isn't theoretical — it's what a real morning-brief pipeline measures every day. The agent's job is to write a script, not to memorise the repo.

ctx_execute runs that script in a sandboxed subprocess (11 supported runtimes), captures stdout, optionally filters with an intent keyword, indexes the full output in FTS5 (so the agent can search it later without re-reading), and returns only the compact summary to the context window.

---

ctx_stats — Usage Aggregation

Aggregate stats across both stats.db and sessions.db. Shows total runs, bytes saved, compression ratios, and session event counts.

{"platform":"cursor","configPath":"/Users/you/.cursor/mcp.json","ok":true,"detail":"would register context-cooler -> ..."}

```

install.py calls this CLI under the hood, one platform at a time.

---

Environment Variables

---

ctx_batch — Multi-Command Pipeline

Run multiple commands and/or search queries in a single MCP call. Each command is executed sequentially with its own intent filter.

ctx_batch(commands=[
  {"language": "python", "code": "...", "intent": "summary"},
  {"language": "shell",  "code": "...", "intent": "top 5"}
], queries=["previous error rates"])

Layer 5: Zero-Token Pipelines

For deterministic tasks, bypass the LLM entirely. Schedule pipelines via launchd/cron:

launchd → python3 pipeline.py → ctx_execute → ctx_deliver → iMessage/Telegram
                              No agent. No model. No tokens.

---