Handle massive contexts (10M+ tokens) with chunking, sub-queries, and free local inference via Ollama
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add massive-context -- uvx massive-context-mcp~/.codex/config.toml
[mcp_servers.massive-context]
command = "uvx"
args = ["massive-context-mcp"]
# codex mcp add massive-context -- uvx massive-context-mcp~/.cursor/mcp.json
{
"mcpServers": {
"massive-context": {
"command": "uvx",
"args": [
"massive-context-mcp"
]
}
}
}.vscode/mcp.json
{
"servers": {
"massive-context": {
"type": "stdio",
"command": "uvx",
"args": [
"massive-context-mcp"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"massive-context": {
"command": "uvx",
"args": [
"massive-context-mcp"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"massive-context": {
"command": "uvx",
"args": [
"massive-context-mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"massive-context": {
"command": "uvx",
"args": [
"massive-context-mcp"
]
}
}
}