검색 및 AI 기억
GitHub에서 내 컴퓨터에 직접 설치해 쓰는 오픈소스 프로그램입니다. 별, 주간 다운로드 수, 뜨는 항목을 볼 수 있습니다.
사용 점수 순: GitHub 스타, 주간 다운로드, 최근 출시 여부, 엔드포인트 응답. 많이 쓴다고 가장 좋은 건 아니에요. 별이 하나도 없이 대량으로 등록한 게시자는 이 탭에 나오지 않습니다. 전체 탭에서 검색하면 찾을 수 있습니다.
카드 보는 법:↓ 주간 다운로드i★ GitHub 스타i세 기둥: 리뷰 · 점검 · 사용 정도i업데이트 = 최신 버전을 레지스트리에 공개한 날i
Velocirag
Lightning-fast RAG for AI agents. 4-layer fusion, ONNX Runtime, sub-200ms search.
Token Compressor
Compress prompts 40-60% using local LLM + embedding validation. Preserves all conditionals.
Web Search
Keyless multi-engine web search, fetch, and clean-Markdown read for AI agents. No API keys.
Gtasks MCP
Manage Google Tasks via MCP: list, search, create, update, move, and clear tasks.
Heor Agent
HEOR MCP server: literature search, CEA, BIA, NMA/MAIC, HTA dossiers (NICE/FDA/EMA/JCA).
Dbgprobe MCP Server
Let AI agents interact with debug probes — connect, flash, read/write memory, and reset.
HackTricks MCP Server
Search and query HackTricks pentesting documentation with quick lookup and section extraction
Mitos
Architectural decision memory for LLM-native workflows — markdown for humans, a graph for agents.
Neuromcp
Semantic memory for AI agents with hybrid search, knowledge graph, and consolidation
Rootz Archive
Search your local Claude Code archive. For continuous capture, use the Rootz Archive plugin.
Memkeeper
Local-first memory for AI agents: on-device hybrid retrieval over a single SQLite file.
TokenKnows
Distill AI coding sessions into weekly reports, ADRs and a knowledge graph - self-hosted
Agg Search
Multi-engine web search with RRF fusion and Jina reranking, plus a tiered fetcher.
GLASTOR pii-veil
Sub-millisecond reversible PII & secret anonymizer in Rust for LLMs, agents, and MCP
Commons
Shared memory, work claims and handoffs so parallel AI coding sessions stop colliding.
Factcovery Facts
500+ short, sourced fun facts: random fact, fact of the day, search and facts by topic.
probbit
A personality, mood and memory layer for AI agents, plus exact plans under hard rules. Offline.
Devlog
Offline developer journal with local-LLM digests, semantic search, and MCP server.
Groundhog
Search, read and research the web, with hidden text stripped by default and sources hashed
Memory Fabric
File-first, local-first MCP memory for AI coding assistants: Markdown + YAML, no vector DB.
llm-sidecar
Local MCP server: web search, cited answers, fact-checking and routed inference. No API key.
N3MemoryCore Lite (Working Memory)
Ephemeral 7-day hybrid vector+BM25 working memory, multilingual, backed by Redis Stack.
Taaghche
Unofficial read-only Taaghche: search Persian ebooks and audiobooks, compare prices, read reviews.