Library
85,762 resources from 11 public sources. 53 checked by hand; the rest are unreviewed.
Results
-
rms-mail-public
MCP serversRMS Mail — High-performance self-hosted email built for large-scale, multi-account workflows.
git clone https://github.com/max-ramas/rms-mail-public.git -
markdown-vault-mcp
MCP serversGeneric markdown vault MCP with hybrid search
git clone https://github.com/pvliesdonk/markdown-vault-mcp.git -
claude-mem-lite
MCP serversPersistent long-term memory for Claude Code via MCP — captures coding decisions, bugfixes, and context across sessions. Hybrid FTS5 + TF-IDF search with episode batching. Single SQLite DB, no external services. Alternative to claude-mem with 600x lower cost.
git clone https://github.com/sdsrss/claude-mem-lite.git -
claudex
MCP serversPersistent memory + FTS5 search MCP server for Claude Code conversation history.
@kunwarshah/claudex@1.3.4 -
mnemon-mcp
MCP serversPersistent layered memory for AI agents. SQLite FTS5, fact versioning, zero-cloud.
mnemon-mcp@1.0.1 -
memory-fts
MCP serversLong-term memory for Claude Code with SQLite FTS5 search and BM25 ranking
claude-memory-fts@1.0.1 -
mismem
MCP serversCooperative persistent memory for LLMs: episodes, memories and traits over SQLite (FTS5, Hebbian)
@lucasmella/mismem@0.2.0 -
Anamnesis
MCP serversCross-machine, file-first memory for Claude Code: markdown + SQLite FTS5, synced with git
anamnesis-memory@0.1.3 -
Srclight
MCP serversDeep code indexing for AI agents. FTS5 + embeddings + call graphs + git intelligence. Fully local.
srclight@0.25.3 -
im-mcp
MCP serversFast, local, read-only MCP Server for WeChat (4.x/3.x) & DingTalk (7.x V2/V3) chat history with FTS5 search
git clone https://github.com/stkoecn/im-mcp.git -
llm-kosh
MCP serversLocal-first AI memory cartridge with persistent MCP memory for Claude via SQLite FTS5.
llm-kosh@2.1.3 -
MCP Memory
MCP serversLocal persistent memory for AI agents: SQLite + FTS5 (BM25) lexical search, zero external infra.
ghcr.io/germaniu/mcp-memory:0.6.0 -
vault-index
MCP serversLocal hybrid keyword + vector search (SQLite FTS5 + sqlite-vec) over markdown vaults and code.
vault-index@0.4.2 -
Memory Kernel
MCP serversLocal-first memory for AI agents: SQLite FTS5, deterministic recall, no vector DB, no cloud.
amormorri-memory-kernel@0.3.2 -
Apple Mail MCP
MCP serversApple Mail MCP server with full-coverage FTS5 body search, reliable on large mailboxes.
apple-mail-mcp@0.5.1 -
LoreConvo
MCP serversLocal-first persistent session memory for Claude Code and other agents. SQLite + FTS5, no cloud.
loreconvo@0.11.0 -
open-memex
MCP serversLocal-first memory for AI coding agents — Markdown + SQLite FTS5, shared across editors via MCP.
open-memex@0.10.0 -
Mac Mail MCP
MCP serversMac Mail MCP server with full-coverage FTS5 body search for local Apple Mail.
mac-mail-mcp@0.5.15 -
Local Memory
MCP serversPersistent local AI memory. SQLite + FTS5 + vector search, knowledge graph. No cloud, no API keys.
@studiomeyer/local-memory-mcp@2.4.4 -
fw-context-mcp
MCP serversBuild-aware code intelligence for embedded C/C++ firmware — libclang, FTS5, call graphs, 37 tools
fw-context-mcp@0.37.0 -
crowbar
MCP serversRetrieve CVEs and index them inside your codebase. One SQLite+FTS5 index, one CLI, one MCP server.
crowbar-cli@0.1.5 -
context-mode
MCP serversFree, local plugin that keeps tool output out of your coding agent's context window. Sandbox and SQLite FTS5 search, 99% fewer bytes in context. Works with Claude Code, Codex, Pi, OMP, OpenCode, Gemini CLI, Copilot CLI and 4 more agents. Zero telemetry.
git clone https://github.com/mksglu/context-mode.git -
KRouter
MCP serversGive your agent a subconscious: mention it and it comes to mind, with what is connected to it, and the lessons carry over. Local plain-Markdown memory for Claude Code, Codex, Cursor and any MCP app. Every answer cites its page. No API key, no server.
git clone https://github.com/398894496-arch/KRouter.git -
CustomsBridge TARIC Engine
MCP serversDeterministic HS6/TARIC customs resolution engine and bilingual FTS5 nomenclature API (UNE Node 04).
Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.