Reduces AI agent token usage by 90% via context compression and task checkpoint persistence.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add smart-context -- npx -y smart-context-mcp~/.codex/config.toml
[mcp_servers.smart-context]
command = "npx"
args = ["-y", "smart-context-mcp"]
# codex mcp add smart-context -- npx -y smart-context-mcp~/.cursor/mcp.json
{
"mcpServers": {
"smart-context": {
"command": "npx",
"args": [
"-y",
"smart-context-mcp"
]
}
}
}.vscode/mcp.json
{
"servers": {
"smart-context": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"smart-context-mcp"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"smart-context": {
"command": "npx",
"args": [
"-y",
"smart-context-mcp"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"smart-context": {
"command": "npx",
"args": [
"-y",
"smart-context-mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"smart-context": {
"command": "npx",
"args": [
"-y",
"smart-context-mcp"
]
}
}
}