DocsContext Gateway

Context Gateway

A transparent proxy that sits between your AI agent and the LLM API, automatically compressing conversation history in the background so you never wait.

View on GitHub

What is Context Gateway?

A transparent proxy that manages context automatically:

  • Monitors token usage across the conversation
  • Pre-computes summaries at 75% of context limit (configurable)
  • Instant compaction when limit is reached — no waiting
  • Compresses large tool outputs on the fly
  • Logs compression events to logs/history_compaction.jsonl

Installation

Quick Install

curl -fsSL https://compresr.ai/api/install | sh

Manual Install Options

  • Download the binary directly from GitHub Releases
  • Build from source (requires Go 1.21+): go build -o context-gateway ./cmd/gateway

Quick Start

1. Launch the interactive wizard

context-gateway

2. Follow the wizard

The TUI wizard will guide you through:

  1. Select your agent (Claude Code, OpenClaw, OpenCode, or Custom)
  2. Enter your LLM provider API key (Anthropic, OpenAI, etc.)
  3. Enter your Compresr API key
  4. Configure compression settings (threshold, model, etc.)

3. Use your agent as usual

The gateway runs as a local proxy on http://localhost:8080. Your agent's API calls are routed through it automatically. No code changes needed.

Supported Agents

Works with any LLM provider (OpenAI, Anthropic, Ollama, Bedrock). Start or stop the gateway anytime — agents auto-detect it.

Claude Code

Codex

OpenHands

OpenClaw

Custom

Usage

# Terminal 1: start the gateway
context-gateway

# Terminal 2: use your agent as usual
claude          # Claude Code
codex           # Codex
openhands       # OpenHands

OpenClaw Integration

# Install the OpenClaw plugin
openclaw plugin install context-gateway

# Start the gateway
context-gateway

# Running or new agents auto-detect the gateway

For custom deployments, point your agent's LLM API base URL to http://localhost:8080.

Configuration

Configuration is saved to ~/.config/context-gateway/.env after running the interactive wizard. You can edit or update your config at any time using the CLI:

Re-configure via CLI

context-gateway -c

Or edit it manually at ~/.config/context-gateway/.env:

Environment Variables

# Required
COMPRESR_API_KEY=cmp_your_api_key        # Your Compresr API key
LLM_API_KEY=sk-xxx                       # Your LLM provider API key

# Agent Configuration
AGENT_TYPE=claude_code                   # claude_code | openclaw | opencode | custom
PROXY_PORT=8080                          # Local proxy port (default: 8080)

# Compression Settings
CONTEXT_THRESHOLD=0.75                   # Trigger compression at 75% of context limit
COMPRESSION_MODEL=espresso_v1            # Model used for history compression
TARGET_COMPRESSION_RATIO=0.5             # How aggressively to compress (0.2-0.9)

# Optional
SLACK_WEBHOOK_URL=https://hooks.slack.com/...  # Slack notifications
LOG_LEVEL=info                           # debug | info | warn | error

Benefits

Zero latency

Compression happens in the background.

Transparent

No code changes needed.

Cost savings

Reduce token usage by 30-70%.

Observable

Full metrics in logs.

Logs & Monitoring

The gateway creates detailed logs for every compression event:

logs/history_compaction.jsonl

When and how conversations are compressed

logs/tool_output_compression.jsonl

Tool output compression metrics and results

logs/telemetry.jsonl

Request/response timing and performance data

Example log entry

{
  "timestamp": "2026-03-06T14:30:00Z",
  "event": "history_compaction",
  "agent": "claude_code",
  "original_tokens": 180000,
  "compressed_tokens": 54000,
  "compression_ratio": 0.7,
  "model": "espresso_v1",
  "latency_ms": 1200
}

Remote Deployment

Deploy the gateway as a service for team-wide usage:

Deploy as a service

# Using Docker
docker run -d \
  -p 8080:8080 \
  -e COMPRESR_API_KEY=cmp_your_api_key \
  -e LLM_API_KEY=sk-xxx \
  -e AGENT_TYPE=claude_code \
  compresr/context-gateway:latest

# Or using the binary directly
COMPRESR_API_KEY=cmp_xxx LLM_API_KEY=sk-xxx context-gateway --port 8080

Environment Variables for Deployment

All configuration options from the ~/.config/context-gateway/.env file can be passed as environment variables. This makes it easy to deploy via Docker, Kubernetes, or any container orchestration platform.

Support