kgLABSCO SUMMARY
Three tools, because the graph language does the work.
Collapsing CRUD into one script runner keeps the tool definitions small while still exposing the whole surface, and pairing it with a live schema call means the model can look up valid relations before it invents one. The trade-off is explicitness: a graph-scoped command is rejected unless you pass graph, use use <YOUR_GRAPH_NAME>, or write the canonical prefix — there is no implicit fallback graph, which is safer but means every script has to say where it is writing.
What it is
A local stdio server over the kg knowledge graph. Architecture, decisions, incidents, rules and dependencies live as typed nodes and edges in *.kg text files you can read, diff and commit, instead of as chunks retrieved from a vector store.
What you get
- Three tools, not thirty:
kg_schemareturns the live schema — valid node types, relations, ID prefixes and edge rules — so the model checks before it writes kg_helpreturns the manual with examples for one domain: node, edge, graph, schema, kql, feedback, batch, script or allkgis the script runner that executes everything else: find, get, add, modify, remove, connect, disconnect, list, stats and audit, several commands per call- IDs read as
<type>:snake_case, and relations are a fixed vocabulary — HAS, USES, TRIGGERS, DEPENDS_ON, DECIDED_BY, READS_FROM and the rest — so the graph stays queryable - Resources expose the working directory and discovered graphs;
kg://graph/{name}returns stats for one graph - Graphs export to an interactive HTML page with
kg graph <YOUR_GRAPH_NAME> export-html
Requirements
The kg-mcp binary, from cargo install kg-cli, the install.sh script, or a GitHub release download, added to your client as a local stdio command. No account and no network service. Graphs default to ~/.kg/graphs; a .kg.toml in the project moves them somewhere you can commit.
Setup effort
Follow the published steps — cargo install kg-cli
USE CASES
01
Building project memory from your existing docs
Point the assistant at a set of files and have it add only facts grounded in them, recording anything ambiguous as a note.
02
Answering questions about how a system actually works
Ask which services write to a datastore or what triggers a retry, and get an answer traversed from stored relations rather than guessed.
03
Keeping the graph honest while you work
Review it for weak descriptions, missing facts and low-value edges, then apply the safe fixes.
04
Reviewing what the assistant remembered
The graph is a text file with readable diffs, so a change to project memory goes through the same review as code.
3 tools
Listed in project docs. Your client may show a different list.
kg_helpkg_schemaBEFORE YOU INSTALL
What to know first
Tools you get
3 tools named in the project's documentation.
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
cargo install kg-cliOr use the install script instead:
TERMINAL
curl -sSL https://raw.githubusercontent.com/nnar1o/kg/master/install.sh | shAdd kg-mcp to your AI client as a local stdio MCP server, with command set to the absolute path of the kg-mcp binary.
Restart your AI client, then check that the kg MCP server appears in its server list.
MCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"kg": {
"command": "<YOUR_PATH_TO_KG_MCP>"
}
}
}
