Live code graph that fuses static analysis (tree-sitter) with OpenTelemetry runtime traces, queried over MCP so AI coding agents get real full-stack context.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add neat -- npx -y @neat.is/mcp~/.codex/config.toml
[mcp_servers.neat]
command = "npx"
args = ["-y", "@neat.is/mcp"]
# codex mcp add neat -- npx -y @neat.is/mcp~/.cursor/mcp.json
{
"mcpServers": {
"neat": {
"command": "npx",
"args": [
"-y",
"@neat.is/mcp"
]
}
}
}.vscode/mcp.json
{
"servers": {
"neat": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@neat.is/mcp"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"neat": {
"command": "npx",
"args": [
"-y",
"@neat.is/mcp"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"neat": {
"command": "npx",
"args": [
"-y",
"@neat.is/mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"neat": {
"command": "npx",
"args": [
"-y",
"@neat.is/mcp"
]
}
}
}