vecscope runs on your own machine — the client starts it, so there is no endpoint to ping. 3 369 installs a week from pypi. Last commit 6 Nov 2026.
Deterministic evidence about embedding spaces and vector indexes: hubness, collapse, ANN recall
Today is the operative word: we check vecscope 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 vecscope -- uvx vecscope
{
"mcpServers": {
"vecscope": {
"command": "uvx",
"args": [
"vecscope"
]
}
}
}
[mcp_servers.vecscope]
command = "uvx"
args = ["vecscope"]
{
"mcpServers": {
"vecscope": {
"command": "uvx",
"args": [
"vecscope"
]
}
}
}
{
"servers": {
"vecscope": {
"type": "stdio",
"command": "uvx",
"args": [
"vecscope"
]
}
}
}
Search an Obsidian vault by tag, term and wikilink graph. Read-only, deterministic, no embeddings.
Local-first memory for AI agents: SQLite FTS5, deterministic recall, no vector DB, no cloud.
AI model releases, price changes and deprecations in one feed: chat, embedding, speech, video.
Create and manage vector database indexes, backups, and collections
Deep code indexing for AI agents. FTS5 + embeddings + call graphs + git intelligence. Fully local.
Operate your Vector Panda collections: search and upsert vectors, schema, measured indexes, exports.
Evaluate RAG retrieval, agent tool calls and prompt regressions with deterministic metrics.
Check content rights before RAG, AI input, training or indexing. Signed RSL evidence, not a license.
Answers built from our own checks of this server.