rastogivaibhav/llm-kosh
latest · 2.1.3Local-first AI memory cartridge with persistent MCP memory for Claude via SQLite FTS5.
- GitHub
- rastogivaibhav/llm-kosh
- PyPI
- llm-kosh
- MCP registry
- io.github.rastogivaibhav/llm-kosh
Install
Needs an env var: CARTRIDGE_WORKSPACE — replace the <placeholders> with your values.
claude mcp add io-github-rastogivaibhav-llm-kosh --env CARTRIDGE_WORKSPACE=<CARTRIDGE_WORKSPACE> -- uvx llm-kosh/mcp confirms it's connected.
Or add it manually in ~/.cursor/mcp.json:
{
"mcpServers": {
"io-github-rastogivaibhav-llm-kosh": {
"command": "uvx",
"args": [
"llm-kosh"
],
"env": {
"CARTRIDGE_WORKSPACE": "<CARTRIDGE_WORKSPACE>"
}
}
}
}In claude_desktop_config.json (Settings → Developer → Edit Config), then restart:
{
"mcpServers": {
"io-github-rastogivaibhav-llm-kosh": {
"command": "uvx",
"args": [
"llm-kosh"
],
"env": {
"CARTRIDGE_WORKSPACE": "<CARTRIDGE_WORKSPACE>"
}
}
}
}Config entry, any MCP client:
{
"command": "uvx",
"args": [
"llm-kosh"
],
"env": {
"CARTRIDGE_WORKSPACE": "<CARTRIDGE_WORKSPACE>"
}
}Community score
Community34/ 100
from web signals · no reviews yet (community v3)
- Web signals (ecosystem baseline)34/100 · 100% of the weight
Used it? Rate v2.1.3
No reviews yet — be the first.
What the web says
Ecosystem baseline34/ 100
Objective public usage & activity (formula v3) — the community score's web-signals input, and its own sort.
GitHubsource ↗
- GitHub stars · 21 days ago
- ★ 1
- Last commit · 21 days ago
- 13 days ago
- Release cadence · 21 days ago
- about every 2 days
- Latest release · 21 days ago
- v1.0.2
- License · 21 days ago
- MIT
Packages
No package-registry data collected yet.
Listed on
- Glama · 1 day ago
- Listed · Glama quality score 2.4 ↗
Mentions
No public discussion scraped yet.
Docs & liveness
No liveness or verification signals yet.
Version history1
- 2.1.3release date unknown
Details
- Transport
- stdio
- Status
- listed
- Source
- Imported from the MCP registry
- Repository
- source ↗