-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathglama.json
More file actions
29 lines (29 loc) · 1.1 KB
/
Copy pathglama.json
File metadata and controls
29 lines (29 loc) · 1.1 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
{
"$schema": "https://glama.ai/mcp/schemas/server.json",
"maintainers": [
"GRU-953"
],
"description": "Local, token-free MCP server that converts any attachment to Markdown and digests it into a deterministic, model-free knowledge-graph memory for any AI assistant — Claude, Gemini, Cursor, VS Code, Windsurf, ChatGPT's Codex and Grok. No LLM, no embedding model, no GPU, no network — runs 100% on-device, byte-identical output, ~0 context tokens to build or recall. Incremental re-digests re-read only what changed; export, import, merge and diff memories across machines. Safe archive unpacking, optional Tesseract OCR, legacy Bengali→Unicode (incl. vetted রম্ন→রু repair), BM25 recall with document/entity-type/multi-project filters, PII suppression. 11 MCP tools.",
"keywords": [
"mcp",
"memory",
"knowledge-graph",
"markitdown",
"ocr",
"deterministic",
"offline",
"token-free",
"graphrag",
"bm25",
"bengali",
"claude",
"gemini",
"cursor",
"vscode",
"windsurf",
"codex",
"grok",
"incremental",
"portable"
]
}