Sinter runs on your own machine — the client starts it, so there is no endpoint to ping. 6 121 installs a week from pypi. Last commit 1 Sep 2026.
Persistent, evidence-backed code graph for coding agents.
Today is the operative word: we check Sinter every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add sinter -- uvx sinter-io
{
"mcpServers": {
"sinter": {
"args": [
"sinter-io"
],
"command": "uvx"
}
}
}
[mcp_servers.sinter]
command = "uvx"
args = ["sinter-io"]
{
"mcpServers": {
"sinter": {
"args": [
"sinter-io"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"sinter": {
"args": [
"sinter-io"
],
"command": "uvx"
}
}
}
Persistent code knowledge graph + 9 composite MCP tools for AI coding agents.
Semantic code graph MCP server for coding agents
Intelligent code knowledge graph for AI coding agents — 71% cheaper, 72% faster
Browser-backed QA with evidence and fix-ready reports for coding agents.
Local proof facts for AI coding clients: intent, tests, Graph Ops, and review evidence.
Code Intelligence Platform for AI Coding Agents
Local code-memory graph for AI coding agents.
Efficient context management: code search, evidence packs, and memory for coding agents.
Answers built from our own checks of this server.