Junê runs on your own machine — the client starts it, so there is no endpoint to ping. 585 installs a week from pypi. Last commit 12 Sep 2026.
Shared, cited knowledge-graph memory for agents — ask, search, remember, keep standing instructions
We read the source, 2 h ago · rules 3dff92dd89df
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
return subprocess.run(["git", "-C", str(root), *args],
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
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 june-mcp -- uvx june-mcp
{
"mcpServers": {
"june-mcp": {
"args": [
"june-mcp"
],
"command": "uvx"
}
}
}
[mcp_servers.june-mcp]
command = "uvx"
args = ["june-mcp"]
{
"mcpServers": {
"june-mcp": {
"args": [
"june-mcp"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"june-mcp": {
"args": [
"june-mcp"
],
"command": "uvx"
}
}
}
This one needs environment variables set before it will start:
JUNE_BASE_URL (Your June endpoint, e.g. http://localhost:8000 (the Junê desktop app's local engine, your own june-local service, or a hosted Team workspace)), JUNE_CANVAS (The canvas (workspace) to bind this connection to — a name (e.g. 'work') or a canvas id; names resolve at startup and ambiguous names fail closed), JUNE_API_KEY (Your June API key (set JUNE_ALLOW_ANON=1 instead to explicitly opt out for keyless local setups)), JUNE_LLM_KEY (Bring-your-own LLM key for cited answers — forwarded per-request as a header, never logged or stored on the service), JUNE_CANVAS_CREATE (Set to 1 to create the named canvas on first run if it doesn't exist yet (refused in read-only mode)), JUNE_READONLY (Set to 1 to hide and refuse all write tools (memory becomes read-only)), JUNE_FILES_ROOT (Opt-in directory agents may upload files from via june_ingest_file — unset means that tool doesn't exist), JUNE_DOCS_CANVAS (Canvas holding the agent's standing docs/skills/learnings (default 'agent_docs')), JUNE_DOCS_REFRESH (Set to 0 to disable the periodic standing-docs digest re-injected on tool results (default on)), JUNE_REFRESH_CALLS (Tool calls between digest injections (default 12) — lower it for small-context agents), JUNE_REFRESH_MINUTES (Wall-clock fallback between digest injections (default 10)), JUNE_DIGEST_CHARS (Standing-docs digest size cap in characters (default 2000)), JUNE_EXPORT_ROOT (Opt-in directory for repo sync — enables june_docs_export / june_page_export / june_page_import (fenced writes, commit-only git); unset means those tools don't exist).
The author declared them in the registry entry; get the values from the project itself.
Persistent memory for AI agents. Remember, recall, and explore a shared knowledge graph.
Persistent memory infrastructure for AI agents. Identity, knowledge graph, and search.
Shared long-term memory for AI agents: save and recall context as a searchable knowledge graph.
Cross-session memory for AI agents - knowledge graph, scoring, semantic search
Event-sourced knowledge graph memory for AI coding agents with semantic search
Semantic memory for AI agents with hybrid search, knowledge graph, and consolidation
Persistent knowledge memory for AI agents. Hybrid search, code graph, pgvector.
Persistent long-term memory for AI agents: semantic search, knowledge graph, and task canvas.
Answers built from our own checks of this server.