24小时 AI 开源项目 实时更新

webui-grambot

Telegram bot that bridges Telegram chats to an Open WebUI instance, letting users chat with LLMs — with per-chat sessions, model selection, image upload and Prometheus metrics.

AI 开源项目版本发布查看详情

agent-engine

Local, read-only dashboard for your AI agents' spend, tokens and activity. Claude Code, Cursor, Codex, OpenCode. No telemetry, no cloud.

AI 开源项目版本发布查看详情

myAI6

Build your own AI assistant: RAG chatbot template with web search, verified citations, conversation compression, and moderation. Ingestion pipeline with intelligent parent-child chunking and vision-based processing of figures, tables, and slides across PDFs, decks, and notebooks (Unstructured). Next.js · Vercel AI SDK · Pinecone.

AI 开源项目版本发布查看详情

promptbranch

Local-first prompt library and version-control system for AI prompts (Desktop App, CLI, MCP Server, and AI execution engine).

AI 开源项目版本发布查看详情

Meridian-AI-Market-Research-and-Strategy-Engine

AI-powered multi-agent market research engine that transforms strategic research briefs into evidence-backed, fully cited consulting-grade reports.

AI 开源项目版本发布查看详情

foreman

Multi-model orchestration engine for AI coding agents: hierarchy, tiers, budgets, gates, dashboard. Director writes tickets; lead/coder/tester/drone execute via opencode.

AI 开源项目版本发布查看详情

ollama-workspace-agent

A Python CLI coding agent powered by remote Ollama, with workspace inspection and command-execution tools over Tailscale.

AI 开源项目版本发布查看详情

claude-dere

Anime-archetype output styles for Claude Code. Same answers, same line numbers, considerably more feelings.

AI 开源项目版本发布查看详情

Athena

Local-first autonomous agent runtime with one durable reasoning loop, governed capabilities, evidence, and universal execution.

AI 开源项目版本发布查看详情

roleiq

Role-specific SME immersion and interview-preparation MVP — JD + resume in, adaptive SME training and interview simulation out.

AI 开源项目版本发布查看详情

codejudge-ai

Deterministic evaluation and reproducible benchmarking for AI-generated code with hardened Docker sandboxing, trusted tests, static analysis, and multi-provider model comparison.

AI 开源项目版本发布查看详情

kaut

KAUT — Knowledge Actualization Under Trust: self-maintaining, AI-first project knowledge for AI agents. Source-bound docs, git-computed freshness verdicts, a gated write path, zero dependencies, MCP server included.

AI 开源项目版本发布查看详情

umbra

Privacy-first multi-model AI workspace: browser-side PII redaction, local-first memory, OpenAI-compatible API.

AI 开源项目版本发布查看详情

colorbox-ai

AI-assisted packaging order review system for color box manufacturers.

AI 开源项目版本发布查看详情

free-deepseek-api

Run DeepSeek Web Chat as a free, local, OpenAI/Anthropic/Responses-compatible API — with tool calling, streaming, reasoning, and per-agent sessions.

AI 开源项目版本发布查看详情

luosida

AI 驱动的工程文档智能审稿系统：LLM + Flask + MySQL，覆盖文本校对/章节完整性/审查要点/工程一致性/历史报告/法规合规/专家意见 7 类自动审查

AI 开源项目版本发布查看详情

ccswitch

Switch Claude Code between your Max subscription and OpenRouter or Groq models, inside the same chat session, and switch back without breaking anything.

AI 开源项目版本发布查看详情

prompt-engineering-evidence

An evidence-graded prompt engineering reference: every technique marked solid / mixed / folklore, every claim sourced, current to 2026. Machine-readable ledger + prompt-injection security + folklore debunked.

AI 开源项目版本发布查看详情

tokenless-web

TokenLess — multi-model AI workspace (frontend, React + Vite). Archived: I built and used this but no longer run the live service. Backend not included.

AI 开源项目版本发布查看详情

MacPulseAI

⚡️ AI-powered macOS process manager & CPU monitor — understand runaway processes with LLM (OpenAI & Anthropic), bilingual UI (中文/English). AI 进程管家

AI 开源项目版本发布查看详情