A local intelligence layer for AI coding agents. Maps your repos, matches the open-source frontier.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add replen --env DIGEST_TOKEN='<DIGEST_TOKEN>' -- npx -y @replen/mcp~/.codex/config.toml
[mcp_servers.replen]
command = "npx"
args = ["-y", "@replen/mcp"]
env = { DIGEST_TOKEN = "<DIGEST_TOKEN>" }
# codex mcp add replen --env DIGEST_TOKEN='<DIGEST_TOKEN>' -- npx -y @replen/mcp~/.cursor/mcp.json
{
"mcpServers": {
"replen": {
"command": "npx",
"args": [
"-y",
"@replen/mcp"
],
"env": {
"DIGEST_TOKEN": "<DIGEST_TOKEN>"
}
}
}
}.vscode/mcp.json
{
"servers": {
"replen": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@replen/mcp"
],
"env": {
"DIGEST_TOKEN": "<DIGEST_TOKEN>"
}
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"replen": {
"command": "npx",
"args": [
"-y",
"@replen/mcp"
],
"env": {
"DIGEST_TOKEN": "<DIGEST_TOKEN>"
}
}
}
}claude_desktop_config.json
{
"mcpServers": {
"replen": {
"command": "npx",
"args": [
"-y",
"@replen/mcp"
],
"env": {
"DIGEST_TOKEN": "<DIGEST_TOKEN>"
}
}
}
}.trae/mcp.json
{
"mcpServers": {
"replen": {
"command": "npx",
"args": [
"-y",
"@replen/mcp"
],
"env": {
"DIGEST_TOKEN": "<DIGEST_TOKEN>"
}
}
}
}Environment variables
DIGEST_TOKEN
Replace the placeholders with your own credentials and keep them out of version control.