llm-grounding 开源项目推荐（17 个）

neuledge /graph

⚡️ Real-time Knowledge Graph for AI Agents. Connect LLMs to verified weather, stock, and currency data via instant tool-calling. No API keys, no scrapers, just grounded facts in <100ms.

JerryLiu369 /agent-web-search

Agent-native multi-provider web search for AI agents (Claude Code, Hermes, DeepSeek Harness, OpenCode, Codex). Model grounding (Responses, ARK) + semantic search (Exa, Parallel) + keyless defaults via MCP, CLI, and DSH plugin.

wirthal1990-tech /USDA-Phytochemical-Database-JSON

76,907 phytochemical records enriched with PubMed, ClinicalTrials.gov, ChEMBL bioactivity & USPTO patents. Production-ready JSON + Parquet. Free 400-row sample. Full dataset: ethno-api.com

openserpapi /mcp

Model Context Protocol server for OpenSERP search engine API

openserpapi /sdk-js

TypeScript / JavaScript SDK for the OpenSERP multi-engine SERP API - Google, Bing, Yandex, Baidu, DuckDuckGo, and Ecosia results in a single call

pfarell /ripple

A weather-radar-style forecast map for lake and river water quality — Sentinel-2, key-free, with a grounded tool-calling assistant

openserpapi /sdk-python

Python SDK for the OpenSERP multi-engine SERP API - Google, Bing, Yandex, Baidu, DuckDuckGo, and Ecosia results in a single call

agarwalujala3-lang /LumenStack-AI

A static codebase analyzer and interactive learning workspace that maps repository architectures with Mermaid and generates source-linked technical tours. Built with Node.js and Express to help developers master unfamiliar codebases, trace dependencies, and practice interview walkthroughs.

soelnvc /GradientRush

Next-gen Cross-Modal Graph RAG engine bridging video keyframes, Whisper speech transcripts, architecture diagrams, and technical PDFs into an interconnected knowledge graph with pgvector and Gemini.

KevinSGarrett /Off_Grid

Evidence-backed commercial intelligence engine that turns construction project data into qualified, CRM-ready sales opportunities for Off Grid Innovation.

FRA-0023 /The-Architecture-of-Opportunity

Data engineering for the creator economy: a three-tier local architecture fusing search intent (Demand) with validated supply metrics to ground generative AI in statistical proof.

beautifulgownss /internal-knowledge-assistant

RAG service with a policy engine that decides whether to answer, ask a clarifying question, or refuse before generation ever runs. Lexical retrieval, SQLite-backed chunk storage, page-level citations, and a mocked generation layer shaped for a drop-in real LLM call.

thpg /conceptuum

A multilingual concept graph with Euler diagrams and executable concept algebra for knowledge exploration, LLM training-data preparation, and evaluation.

openserpapi /n8n

n8n community node for OpenSERP. It supports web search, image search, URL extraction + Cloud account/pricing calls and engine capability/status calls.

Angelica-M /ai201-project1-rag-system-with-guardrails-v2026

CLI RAG system (with ChromaDB vector database) that consists of a relevance-gate guardrail against hallucination, grounded source-cited answers, and acceptance criteria validated through repeated, reproducible test runs for each of 5 success criteria rather than a single pass.

pulkitj /groundguard

Verify LLM output against your source documents. Catch hallucinations in RAG pipelines and agentic workflows before they reach users.

automdev /web-search-api

Web Search API by Autom: One call, several search engines, first organic results. Python, TypeScript, cURL, PHP, C#, Go, Java, Ruby examples.