Vibe Coding DiscoverCategory
AI Tools
Developer tools for evaluation, observability, and AI operations.
| Project name | Stars | Category | Language | Tags | Summary | View Project |
|---|---|---|---|---|---|---|
| AnyJev | ★ 323 | AI Tools | Python | calibration · decision-making · llm · logits · uncertainty-quantification · training-free · probabilities · typed-decisions · system-one · inference · evaluation · temperature-scaling · hidden-states · human-in-the-loop · qwen3 · probability-thresholding | Python library that turns an open LLM into a calibrated decision model by reading typed questions from one next-token prefill. L0 removes position/label bias with zero labels; L1/L2 add temperature or closed-form heads for real probabilities, with shipped Qwen3 heads. | View Project → |
| StatsPAI | ★ 322 | AI Tools | Python | agent-native · mcp · causal-inference · econometrics · statistics · skills · python · difference-in-differences · instrumental-variables · panel-data · double-machine-learning · synthetic-control · regression-discontinuity · treatment-effects · stata · rstats | Agent-native Python library for causal inference and applied econometrics, offering a unified Stata/R-style API, machine-readable function schemas, a bundled MCP server, and Skills for LLM agents. Covers DiD, IV, RD, synthetic control, matching, DML, and panel models with R/Stata parity checks. | View Project → |
| scribe.js | ★ 320 | AI Tools | JavaScript | ocr · text-extraction · pdf · tesseract · webassembly · document-ai · searchable-pdf · mcp · cli · browser · nodejs · image-processing · table-extraction · form-fields · javascript-library | JavaScript OCR and text-extraction library for images and PDFs, powered by WebAssembly Tesseract with no build step. Runs in browser or Node, outputs searchable PDFs, and ships a CLI plus an MCP server for agent/tool integration. | View Project → |
| docjev | ★ 308 | AI Tools | Python | document-classification · document-splitting · ocr · pdf · llm-inference · cli · python · llamaparse · document-ai · benchmarking · pdf-parsing · docx · pptx · fastapi · pydantic · page-boundaries | DocJev is a Python library, CLI, and local web app that classifies PDFs/DOCX/PPTX into your own natural-language categories and splits multi-document packets into page ranges. It uses local LiteParse OCR plus hosted Jev (or an OpenAI baseline) decisions, with review flags, PDF export, and a benchmark harness. | View Project → |
| bg0 | ★ 304 | AI Tools | TypeScript | background-removal · browser-inference · webgpu · wasm · onnx · privacy-preserving · image-editing · local-first · npm-package · biRefNet · transparent-png · no-backend · typescript · monorepo · react | BG0 is an open-source background remover that runs fully in the browser using a BiRefNet ONNX model over WebGPU with a WASM fallback, so images never leave the device. It ships a TanStack Start web app plus a reusable @bg0/browser package with progress, cancellation, and transparent PNG output. | View Project → |
| backlot | ★ 295 | AI Tools | Python | mcp · mock-server · api-emulation · rag-evaluation · agent-testing · deterministic-corpus · acl · pagination · official-sdks · fastapi · agent-skills · llamaindex · slack · gmail · jira · s3 | Local emulator of enterprise SaaS APIs (Slack, Gmail, Drive, GitHub, Jira, Notion, S3) serving vendor-shaped responses, pagination, auth and per-document ACLs from a deterministic corpus you supply. Ships an MCP server and agent skill for offline agent and RAG evaluation. | View Project → |
| kitaru | ★ 292 | AI Tools | Python | agent-evals · observability · replay · tracing · llmops · regression-testing · self-hosted · mcp · durable-execution · checkpoints · ci-cd · session-recording · human-in-the-loop · python · typescript · open-source | Kitaru records or imports production AI agent traces as sessions, then replays the real agent code against changes in model, prompt, or tools so you see what improved or regressed before shipping. Self-hosted, Apache 2.0, with Python/TypeScript SDKs, framework adapters, MCP server, and agent skills. | View Project → |
| emulo | ★ 291 | AI Tools | HTML | agent-personalization · agent-memory · session-mining · local-first · Claude Code · Codex · MCP · agent-skills · privacy | Emulo mines local AI coding session logs into a personal profile that agents can load before tasks. It supports Claude Code, Codex, and adapters for other coding agents, with CLI, plugins, and an MCP server. | View Project → |
| freebucks-proxy | ★ 291 | AI Tools | Go | openai-compatible · api-gateway · llm-router · proxy · multi-account · token-pool · self-hosted · session-management · tls-fingerprinting · stealth · admin-dashboard · single-binary · anthropic-compatible · golang · sqlite · coding-agent | Go gateway exposing OpenAI- and Anthropic-compatible /v1 endpoints in front of a coding-model provider, pooling multiple account tokens with session lifecycle and credit metering, optional browser-like TLS stealth, and an embedded Svelte admin dashboard. Single binary, self-hosted, no CLI. | View Project → |
| jev-chat-jarvis-mac | ★ 285 | AI Tools | Python | macos · wechat · ocr · local-first · privacy · intent-recognition · reply-generation · screen-capture · pyobjc · desktop-assistant · risk-scoring · read-only · llm · on-device-inference | macOS 悬浮窗：只读抓取微信聊天窗口并 OCR，用本地 decider-2b 判断消息意图与风险分级，再经 LLM 并发生成候选回复并按本地排序重排，可一键填入微信输入框；不注入、不自动发送、数据可全本地。 | View Project → |
| engrim | ★ 282 | AI Tools | Python | episodic-memory · memory · local-first · sqlite · mcp · context-management · cross-model · ai-agents · claude-code · cursor · codex-cli · opencode · windsurf · antigravity · hybrid-search · developer-tools | Engrim is a local-first SQLite episodic memory engine for AI coding agents. It injects a ~4k-char memory pack at session boot and shares one project-scoped store across Claude Code, Cursor, Codex CLI, OpenCode, Copilot CLI and Antigravity via hooks plus an MCP server. | View Project → |
| OrcaReplay | ★ 268 | AI Tools | TypeScript | agent-debugging · agent-tracing · llm-observability · replay · trace-forking · deterministic-replay · proxy-capture · mcp · llm-evaluation · checkpoints · cli · typescript · apache-2.0 · offline-replay · model-comparison · tls-intercept | OrcaReplay records AI agent runs below the agent — at the proxy, shell, MCP and filesystem boundary — then replays them offline byte-for-byte and forks them at any checkpoint onto a different model. CLI tool for debugging why an agent run did what it did. | View Project → |
| ConsistI2V | ★ 261 | AI Tools | Python | image-to-video · video-generation · diffusion-models · generative-ai · visual-consistency · spatiotemporal-attention · video-synthesis · pytorch · gradio · research-paper · tmlr-2024 · text-prompted-video · first-frame-conditioning · inference · training | Official code for ConsistI2V (TMLR 2024), a diffusion method that turns a still image plus text prompt into a consistent video using first-frame spatiotemporal attention and low-frequency noise initialization. Includes inference and training scripts, a Gradio demo, and Hugging Face weights. | View Project → |
| jentic-public-apis | ★ 259 | AI Tools | — | API catalog · AI agents · OpenAPI · Arazzo · API discovery · API workflows · machine-readable · AI readiness | A community catalog of public API specifications and Arazzo workflows designed to give AI agents structured API knowledge. Includes machine-readable OpenAPI files and supports AI-readiness scoring. | View Project → |
| safelabs-eval | ★ 257 | AI Tools | Python | red-teaming · agent-security · owasp-asi · prompt-injection · jailbreak-detection · llm-security · evaluation · ai-safety · adversarial-prompts · security-testing · benchmark · cli · ci-cd · detectors · python · guardrails | safelabs-eval is a framework-agnostic Python red-teaming toolkit that fires 131 OWASP ASI-aligned adversarial prompts at any agent HTTP endpoint or Python callable, scoring responses with pure-Python detectors (no LLM cost) and emitting structured JSON reports for CI. | View Project → |
| BeefTV | ★ 256 | AI Tools | TypeScript | AI video creation · creative workspace · node-based canvas · image generation · video generation · audio generation · AI agents · local-first · multimodal workflows | A local-first AI video and media creation workspace with a node-based canvas for connecting prompts, assets, models, and agents. Supports configurable model providers and workflows for generating and editing text, images, video, and audio. | View Project → |
| Prism-Legal-OS | ★ 253 | AI Tools | TypeScript | legal AI · legal assistant · contract analysis · legal research · document drafting · document review · RAG · self-hosted | Self-hostable legal workspace with an AI assistant for document drafting, contract review, and legal research. It supports source-backed document search using Qdrant and lets users configure OpenAI, Anthropic, Google, or OpenAI-compatible providers. | View Project → |
| plate-playground-template | ★ 240 | AI Tools | TypeScript | rich-text-editor · plate · nextjs · react19 · tailwindcss · shadcn-ui · mcp · ai-editor · slate · ai-sdk · radix-ui · template · typescript · slash-command · docx · markdown | Next.js 16 + React 19 starter template for a Plate rich-text editor with AI plugins, shadcn/ui components, and MCP support. Uses the Vercel AI SDK and a user-supplied AI Gateway key to power inline AI editing commands. | View Project → |
| gflow-cli | ★ 237 | AI Tools | Python | cli · mcp · mcp-server · video-generation · image-generation · veo · imagen · google-flow · text-to-video · image-to-video · batch-generation · playwright · browser-automation · generative-ai · alpha · unofficial | Python CLI plus MCP server that drives Google Flow's Veo and Imagen models from the terminal for scripted, batched text-to-video, image-to-video and text-to-image work. Alpha, reverse-engineered via headed Playwright and unofficial; install the Claude plugin or MCP server so agents can generate too. | View Project → |
| agent-inspect | ★ 233 | AI Tools | TypeScript | observability · debugging · tracing · trajectory-testing · local-first · cli · ci · jsonl · evidence-bundle · agent-debugging · devtools · typescript · redaction · mcp · testing · deterministic-checks | Local-first TypeScript toolkit for AI agent observability: turns JSONL runs into execution trees, runs deterministic trajectory checks in CI, and exports redacted hash-verifiable evidence bundles. Includes framework adapters (LangChain, LangGraph, AI SDK, OpenAI Agents) and an optional read-only MCP server. | View Project → |
| bitrouter | ★ 231 | AI Tools | Rust | llm-router · model-routing · llm-gateway · mcp · acp · cost-optimization · observability · opentelemetry · guardrails · agent-harness · openai-compatible · anthropic-compatible · policy-lock · rust · cli · local-proxy | Rust LLM router and gateway that sits in front of any agent harness: routes each model call, MCP tool/skill, and ACP sub-agent to the cheapest adequate path, with OTel telemetry and a git-owned policy-lock it improves from admitted eval evidence. Reports ~33% cost cut on Terminal-Bench 2.1 at near-parity accuracy. | View Project → |
| AgentMeasure | ★ 218 | AI Tools | Python | agent-observability · token-accounting · telemetry · conformance-testing · llm-metrics · usage-tracking · opentelemetry · local-first · agent-evaluation · mcp · codex · claude-code · ci-checks · benchmarking · standards · quality-ladder | Open measurement and conformance layer for AI-agent telemetry. Audits token and usage accounting with PASS/FAIL/UNPROVABLE CI checks, reads Codex rollout logs locally to expose retries and duplicate records, and runs preregistered harness experiments. | View Project → |
| node9-proxy | ★ 216 | AI Tools | TypeScript | agent-security · mcp-security · prompt-injection · dlp · audit-logs · llm-security · guardrails · human-in-the-loop · devsecops · claude-code · github-actions · ssrf · governance · sandboxing · cli · security-proxy | Node9 is a security proxy that sits between AI coding agents and their tools: it scans past sessions for leaks, blocks risky commands before execution, gates MCP servers, and produces audit reports. Works with Claude Code, Codex, Cursor and any MCP server. | View Project → |
| design-studio-ai | ★ 215 | AI Tools | TypeScript | agent-first · design-workspace · mcp · webmcp · byok · cli · agent-skill · 3d · motion · slides · wireframes · exports · self-hosted · react · collaboration · design-systems | Agent-first design workspace where humans and LLM agents co-edit web layouts, slides, wireframes, 3D scenes, and motion videos on one versioned document. Exposes MCP/WebMCP, a dsa CLI and an agent skill, with BYOK media generation and multi-format export. | View Project → |
| inference-gateway | ★ 209 | AI Tools | Go | llm-gateway · proxy · openai-compatible · multi-provider · mcp · observability · kubernetes · self-hosted · go · opentelemetry · streaming · function-calling · multimodal · oidc · docker · provider-routing | A Go, cloud-native gateway that unifies many LLM providers (OpenAI, Anthropic, Groq, Ollama, DeepSeek, and more) behind one OpenAI-compatible API. Adds MCP tool discovery, streaming, vision, OIDC auth, and OpenTelemetry/Prometheus metrics for self-hosted LLM traffic. | View Project → |
| storycast | ★ 206 | AI Tools | TypeScript | AI video generation · storytelling · animated films · text-to-video · AI images · text-to-speech · lip sync · subtitles · multilingual | A browser-based AI film maker that turns a topic and chosen character into a narrated, animated short film. It generates scripts, images, video shots, voice, music, edits, and subtitles using fal.ai; users provide their own fal key. | View Project → |
| ETOS-LLM-Studio | ★ 205 | AI Tools | Swift | ios · watchos · swift · swiftui · local-llm · llama.cpp · gguf · offline-inference · mcp · rag · agent · siri-shortcuts · on-device-embeddings · multimodal · speech · cross-device-sync | Native iOS/watchOS LLM client in Swift: runs local GGUF models offline via llama.cpp or connects to OpenAI, Claude and Gemini. Adds session-level Chat/Agent modes, MCP tools, Agent Skills, an embedded Alpine Linux sandbox, Browser Agent, local SQLite RAG memory and two-way device sync. | View Project → |
| proxy | ★ 204 | AI Tools | TypeScript | llm-proxy · cost-tracking · model-routing · observability · budget-caps · kill-switch · anomaly-detection · claude-code · local-first · llm-gateway · cost-attribution · failover · agent-ops · openai-compatible · anthropic-compatible · sqlite | Local-first MIT proxy that sits between AI agents and model providers, pricing every request and rolling cost up per run, agent, and model. Adds budget caps, a kill switch, anomaly detection, complexity-based model routing, and cross-provider failover behind one Anthropic/OpenAI-compatible base URL. | View Project → |
| Token-Print | ★ 203 | AI Tools | TypeScript | transformer-visualization · llm-interpretability · mechanistic-interpretability · 3d-visualization · threejs · webgl · gguf · attention-heads · forward-pass · activation-debugging · rope · gqa · model-inspector · websocket · huggingface | TokenPrint renders a real LLM forward pass in interactive 3D: attention heads, RoPE, GQA and the residual stream traced from actual tensors or a local GGUF file. Four modes (architecture, generation, walkthrough, debugger) with provenance tagging. Runs a FastAPI backend plus a Next.js/Three.js frontend. | View Project → |
| awesome-jev | ★ 202 | AI Tools | Python | awesome-list · Jev · typed-decisions · AI-agents · agent-skills · browser-automation · computer-use · MCP · coding-agents · SDKs · curated-directory | A curated directory of Jev demos, SDKs, skills, and integrations, with source links, recorded examples, and notes on setup and limitations. Covers browser and computer-use agents, MCP tools, coding-agent integrations, and other typed-decision applications. | View Project → |
| eGEOagents | ★ 194 | AI Tools | Python | generative engine optimization · answer engine optimization · AI SEO · LLM SEO · content optimization · AI search visibility · citation tracking · CLI · Claude Code · MCP · JSON-LD · llms.txt | Python toolkit and CLI for auditing and optimizing web content for citations in AI search engines. Includes a content analysis, ranking, rewriting, and schema-generation pipeline, plus Claude Code skills, MCP support, and an evaluation harness. | View Project → |
| ThinkingOrbs | ★ 192 | AI Tools | Swift | SwiftUI · loading indicators · AI interfaces · agent interfaces · animations · iOS · accessibility | A SwiftUI package of nine animated, dotted 3D indicators and status labels for AI and agent interfaces. Supports state-specific designs, adjustable sizing, accessibility features, and pausing when offscreen or in the background. | View Project → |
| pad | ★ 183 | AI Tools | Go | AI agents · project management · MCP server · agent skills · CLI · local-first · task management · self-hosted · SQLite | Local-first project management app with a CLI, web UI, MCP server, and skills for coding agents. Agents can manage tasks and project knowledge while their actions are recorded in an activity feed. | View Project → |
| www-project-agent-memory-guard | ★ 183 | AI Tools | Python | agent-memory · ai-security · prompt-injection · guardrails · memory-poisoning · llm-security · owasp · ai-safety · agentic-ai · policy-enforcement · rag-security · middleware · forensics · mcp-server · opentelemetry · python | OWASP Agent Memory Guard is a Python runtime defense layer that sits between an AI agent and its memory store, screening writes/reads for prompt injection, memory poisoning, secret leakage, protected-key tampering and self-reinforcement loops. It enforces YAML policies, emits audit events and ships drop-in integrations | View Project → |
| dynolab | ★ 178 | AI Tools | Swift | ai-safety · interpretability · mechanistic-interpretability · alignment · activations · probes · sparse-autoencoders · activation-patching · steering · local-inference · apple-silicon · mlx · macos · swiftui · python-sdk · mcp | A free macOS Apple Silicon workbench for AI safety and interpretability research. Run local MLX models, capture activations, train probes and SAEs, run interventions, and script or automate studies via a Python SDK, HTTP API and MCP bridge. | View Project → |
| crawlberg | ★ 178 | AI Tools | Rust | web-crawling · web-scraping · markdown · mcp · mcp-server · rust · multi-language-bindings · ffi · headless-browser · rag-data · json-ld · llm-context · cli · wasm · sitemap · robots-txt | Rust web-crawling engine that turns sites into clean Markdown and structured metadata, with 14 language bindings plus an MCP server and agent skills for feeding AI agents and RAG pipelines. SSRF-safe, robots-aware, optional headless-browser rendering. | View Project → |
| AnvilWiki | ★ 173 | AI Tools | TypeScript | astro · cloudflare-pages · game-wiki · static-site · mdx · tailwindcss · seo · i18n · mcp · agent-skills · pagefind · content-pipeline · adsense · lighthouse-100 · wiki-template · typescript | Astro + Cloudflare Pages template for SEO-driven game wiki sites: static, zero-JS, Lighthouse 4x100, AdSense slots, i18n. It bundles AI content skills and an MCP ops tool, but the product itself is a website template, not an AI system. | View Project → |
| remote-agent-browser | ★ 169 | AI Tools | TypeScript | browser automation · computer use · cloud browser · agent browser · sandbox · Chromium · remote browser | A TypeScript client that runs agent-browser in disposable Vercel Sandboxes. It provides persistent browser sessions, command execution, screenshots, file transfer, and proxy support for agent-driven web automation. | View Project → |
| AiNiee-Next | ★ 167 | AI Tools | Python | translation · llm · batch-processing · cli · tui · mcp-server · rag · plugins · multi-model · async · uv · epub · subtitles · ocr · web-ui · game-localization | CLI/TUI batch translation fork of AiNiee powered by LLM APIs (OpenAI, Anthropic, Gemini, DeepSeek, local models). Handles 20+ file formats at 100+ async concurrency, with profiles, task queue, RAG glossary reference, plugins, web dashboard and an optional MCP server. | View Project → |
| llm-api-setup-guides | ★ 164 | AI Tools | Python | chinese · tutorials · documentation · llm-api · api-relay · openai-compatible · byok · claude-code · codex-cli · cursor · cherry-studio · local-llm · rag · mcp · troubleshooting · mkdocs | Chinese-language documentation hub (98 guides) for wiring LLM APIs into tools: base URL/key setup for chat clients, coding agents (Claude Code, Codex, Cursor, Cline), local model servers and RAG platforms, plus billing checks and 401/404/429 troubleshooting. | View Project → |
| claude-s40 | ★ 157 | AI Tools | Java | Claude · chat client · Java ME · Nokia Series 40 · web search · legacy mobile | A Java ME Claude chat client for Nokia Series 40 phones, backed by a small Go server that handles modern Claude API requests, web search, pairing, and chat storage. Includes a custom TLS setup for older phones. | View Project → |
| toon-java | ★ 155 | AI Tools | Java | TOON · serialization · LLM contexts · token reduction · encoder · decoder · Java · structured data | JToon is a Java library for encoding and decoding TOON, a compact structured-data format for LLM contexts. It supports Java objects, JSON input, Jackson annotations, configurable delimiters, and strict decoding. | View Project → |
| context-engineering | ★ 155 | AI Tools | Python | context-engineering · llm · ai-agents · mcp · rag · memory-management · multi-agent · prompting · agent-skills · evals · observability · orchestration · book-companion · examples · generative-ai · tutorial | Companion repo for a book on context engineering: runnable Python/JS/Java/Jupyter examples covering RAG, agent memory, tools and MCP servers, multi-agent orchestration, prompting patterns, evals, and governance. A hands-on reference for building reliable LLM systems. | View Project → |
| goutoujunshi-jev-chat | ★ 152 | AI Tools | Python | chat assistant · reply drafting · OCR · WeChat · desktop overlay · conversation analysis · relationship trends · human-in-the-loop · macOS · Windows · Android | Cross-platform chat assistant that reads visible conversations, analyzes context, and drafts ranked replies in a floating window. Users review OCR and suggestions before copying or filling a draft; sending remains manual. Android currently cannot capture WeChat chats. | View Project → |
| aiworkdeck | ★ 150 | AI Tools | Java | ai-workspace · legaltech · document-ai · ai-agents · mcp · rag · ocr · self-hosted · electron · spring-boot · vue · plugin-marketplace · tracked-changes · due-diligence · desensitization · text-to-speech | Self-hosted AI-native document workspace for legal teams: agents, MCP tool orchestration, plugin/skill marketplace, embedded LibreOffice editing with tracked changes, OCR, RAG, and evidence-chain review. Java/Spring backend, Vue + Electron desktop, optional local models via Ollama. | View Project → |
| servo-fetch | ★ 149 | AI Tools | Rust | web-scraping · headless-browser · markdown-extraction · mcp-server · agent-skills · rust · cli · screenshots · crawling · sitemap · javascript-execution · accessibility-tree · json-extraction · browser-engine · http-api · python-node-sdk | Rust browser engine built on embedded Servo that renders pages and extracts Markdown, schema-driven JSON, or PNG screenshots with no Chromium or API key. Ships as CLI, Rust/Python/Node libraries, HTTP API, built-in MCP server, and agent skill for feeding agents web content. | View Project → |
| WebReaper | ★ 148 | AI Tools | C# | web scraping · web crawling · LLM extraction · Markdown · MCP server · Claude Code skill · structured data · browser automation · C# | WebReaper is a self-hosted .NET scraper with a CLI and library for crawling sites, producing Markdown, and extracting structured data with optional LLM support. It also provides MCP servers and a Claude Code skill for agent-driven scraping. | View Project → |
| canonry | ★ 145 | AI Tools | TypeScript | aeo · geo · ai-visibility · answer-engine-optimization · seo · brand-monitoring · mcp · cli · self-hosted · local-first · sqlite · web-analytics · server-side-tracking · agent-first · competitor-analysis · byok | Self-hosted, agent-first AEO/GEO platform (CLI, MCP, API, web UI) that tracks brand mentions and citations across AI answer engines, crawls sites for technical SEO health, and runs competitor visibility analysis on SQLite with BYOK provider keys. | View Project → |
| open | ★ 144 | AI Tools | Go | AI visibility · AEO · GEO · LLMO · brand monitoring · citation analysis · share of voice · self-hosted · MCP · SQLite | Self-hosted tool for tracking brand mentions, rankings, and citations across AI search engines. Includes a dashboard, scheduled evaluations, SQLite storage, data export, and an MCP server. | View Project → |
| dasheng | ★ 144 | AI Tools | JavaScript | streaming-asr · speech-recognition · pronunciation-scoring · llm-judge · real-time · self-hosted · text-alignment · english-learning · websocket · pcm16 · voice · nextjs · audio | A Next.js web app that scores English read-aloud: streaming R2T2 ASR transcribes speech, word-level alignment finds mismatches, and the Jev LLM answers closed questions (same word? error type? same sentence?) to flag errors and compute a weighted total. Self-hostable on a 12GB NVIDIA GPU or via a relay demo; pronunciat | View Project → |
| shellward | ★ 136 | AI Tools | TypeScript | ai-security · prompt-injection · dlp · guardrails · pii-detection · mcp-security · data-exfiltration · llm-security · ai-firewall · compliance · china-pipl · zero-dependency · cli · agent-skill · mcp-server · audit-log | ShellWard is a zero-dependency AI agent security and China-compliance gateway. It scans projects for cross-border data flow, hardcoded secrets and Chinese PII (PIPL/CSL/MLPS), then blocks prompt injection, data exfiltration and dangerous tool calls at runtime via SDK, MCP server or CLI. | View Project → |
| vellium | ★ 134 | AI Tools | TypeScript | local-first · desktop-app · AI-chat · roleplay · AI-writing · RAG · MCP · local-LLMs · live-voice · plugins · OpenAI-compatible | A local-first desktop AI workbench for chat, character roleplay, and long-form writing. It supports OpenAI-compatible providers and local model backends, with SQLite storage, RAG collections, MCP tools, plugins, and live voice features. | View Project → |
| aikito | ★ 133 | AI Tools | Python | agent-memory · agent-governance · agent-skills · mcp · subagents · cli · coding-agents · multi-agent · workspace-sync · git-based · portable-config · durable-memory · python · stdio-only · cross-platform | Aikito is a Python CLI that keeps coding-agent instructions, skills, MCP definitions, subagents and durable Markdown memory in one Git-managed workspace synced across agents, projects and machines. Stdlib-only, no daemon or vector store; adds adopt/sync/status/doctor commands with preflight safety checks. | View Project → |
| knowclip | ★ 128 | AI Tools | TypeScript | video-editing · speech-recognition · ASR · highlight-extraction · ffmpeg · local-inference · LLM · desktop-app · subtitles · short-video · privacy · batch-export · Tauri · Paraformer | KnowClip is a Tauri + React + FastAPI desktop app that transcribes long videos locally with Paraformer, uses your own OpenAI-compatible LLM key to pick highlight segments, and batch-exports clips via ffmpeg stream copy without re-encoding. | View Project → |
| protocol-launcher | ★ 127 | AI Tools | TypeScript | deep-link · url-scheme · mcp-installer · one-click · llm-provider · cursor · cherry-studio · vscode · codex · opencode · typescript · esm · tree-shaking · developer-tools · url-generator · zero-dependencies | Type-safe TypeScript library that builds deep links and URL scheme links for protocol-based apps, including one-click MCP server installs for Cursor and provider imports for Cherry Studio. Useful for wiring AI tool launch/install buttons into docs, sites, or CLIs. | View Project → |
| codified-prompt-rule-engine | ★ 123 | AI Tools | HTML | prompt engineering · prompt optimization · meta-prompting · system prompts · few-shot prompting · Claude · LLMs | A static HTML project describing a prompt-optimization pipeline that adds task structure, constraints, confidence boundaries, and self-check instructions. The repository appears to contain a landing page rather than an implemented optimization engine. | View Project → |
| awesome-ai-gateway | ★ 119 | AI Tools | HTML | AI gateways · LLM routing · LLM proxy · benchmarks · cost optimization · self-hosted · MCP · security · compliance · awesome list | A bilingual catalog and comparison guide to 160+ AI gateways, with decision tools, reproducible cost benchmarks, security and compliance data, and an eight-chapter handbook. Helps developers choose and evaluate gateways for LLM routing and agent traffic. | View Project → |
| claude-gateway-manager | ★ 119 | AI Tools | HTML | Anthropic API · Claude · API proxy · LLM gateway · model routing · account pooling · rate limiting · token budgets · failover | Describes a gateway for routing Anthropic Messages API requests across Claude models and account pools, with budget controls, failover, and monitoring. The repository appears to contain a static HTML landing page rather than the gateway implementation. | View Project → |
| dsh-config-manager | ★ 119 | AI Tools | TypeScript | backup · restore · migration · sync · config-manager · dsh-plugin · deepseek-harness · webdav · git-sync · encryption · snapshot · marketplace · cli · mcp · skills · typescript | A DeepSeek Harness (DSH) plugin that exports, imports, migrates, schedules and syncs your complete DSH setup — settings, plugins, MCP servers, skills, agent presets and workspaces — with secret stripping, dry-run preview, rollback and Git/WebDAV remote sync. | View Project → |
| Miffan | ★ 118 | AI Tools | Kotlin | android · ai-chat-client · byok · multi-provider · mcp · skills · mcp-client · local-first · kotlin · jetpack-compose · ssh-workspace · selected-text-translation · tts-stt · character-cards · self-hosted · rikkahub-fork | Miffan is an open-source Android AI client (RikkaHub fork) that connects OpenAI, Gemini, Claude and compatible endpoints with BYOK or ChatGPT Codex sign-in. Each assistant has isolated prompts, memory, MCP servers, Skills and tools, plus local Linux or SSH remote workspaces, TTS/STT and selected-text translation. | View Project → |
| Skill2Env | ★ 115 | AI Tools | Python | agent evaluation · RL tasks · task generation · terminal agents · Agent Skills · benchmarking · Docker | Skill2Env converts Agent Skills into Docker-based, host-validated terminal-agent tasks in Harbor format, ready for reinforcement learning or dataset publication. It uses Codex agents to plan and create tasks. | View Project → |
| OGAD | ★ 115 | AI Tools | TypeScript | local-ai · on-device · privacy · offline · openai-compatible · llama.cpp · gguf · image-generation · speech-to-text · text-to-speech · mcp · rag · electron · desktop-app · vision · local-first | Off Grid AI is a local-first desktop runtime and studio that runs open models (text, vision, image, voice) on-device via llama.cpp, stable-diffusion.cpp, whisper.cpp and Kokoro, exposed through one OpenAI-compatible gateway. Includes chat, RAG projects, MCP connectors, artifacts, and an opt-in Pro memory layer. | View Project → |
| agentburn | ★ 114 | AI Tools | Python | usage tracking · cost profiling · token usage · observability · rate limits · local-first · CLI · agent analytics · MCP | A local, read-only CLI profiler for AI coding agents and assistants. It analyzes usage logs to show peak limits, token costs by source, overnight spending, and behavioral causes such as retry loops, then suggests configuration fixes. | View Project → |
| swarm-orchestrator | ★ 113 | AI Tools | JavaScript | code verification · AI-generated code · CI · testing · evidence ledger · provenance · tamper-evident · GitHub Actions · coding agent | Swarm Verify runs repository checks and records tamper-evident evidence in verifiable bundles. It integrates with GitHub Actions to report what was tested and what remains unverified; the repository also includes an optional beta coding agent. | View Project → |
| Heisenberg | ★ 112 | AI Tools | PHP | laravel · php · cms · blog-engine · block-editor · gutenberg · media-library · bilingual · i18n · mcp · ai-writing-assistant · livewire · roles-permissions · email-templating · headless-cms · content-engine | Laravel package providing a Gutenberg-style block editor, content model, media library, templates and bilingual blog backend with zero host coupling. Ships an optional AI writing assistant and MCP server, but its core is a CMS, not an AI/agent product. | View Project → |
| Hermes-router | ★ 110 | AI Tools | Python | LLM router · API proxy · provider failover · key rotation · model routing · caching · usage analytics · rate limiting · local models · OpenAI-compatible · Anthropic-compatible · embeddings · tool calling · Prometheus | A self-hosted Python gateway for OpenAI- and Anthropic-compatible requests. It routes across cloud and local model providers with failover, key rotation, caching, budgets, and a monitoring dashboard. | View Project → |
| easyedit | ★ 107 | AI Tools | Python | video-generation · video-editing · fan-edits · captions · beat-detection · local-first · Claude Code · Codex · agent-friendly · Whisper | Local-first Python tool that turns a movie title into a captioned speech and beat-cut montage. It uses Claude Code or Codex for planning, faster-whisper for transcription, and HyperFrames for rendering; it can also run with heuristic defaults. | View Project → |
| guardana | ★ 101 | AI Tools | HTML | AI security · LLM security · agent security · MCP security · red teaming · model scanning · prompt injection · supply chain security · CI/CD · SARIF · OpenTelemetry · security verification | Python security verification tool for AI artifacts, live model endpoints, MCP servers, and recorded agent traces. Runs offline artifact scans or bounded active probes, produces reproducible findings, and can gate CI with SARIF output. | View Project → |
| know-where | ★ 101 | AI Tools | Python | knowledge management · AI summarization · content archiving · second brain · Feishu · Lark bot · web extraction · multimodal · speech transcription · self-hosted | A self-hosted Feishu bot and CLI that extracts public content from supported links, uses configurable AI models for classification and summaries, and archives results in Feishu Bitable. Supports articles, GitHub READMEs, image posts, and video transcription. | View Project → |
| Magic-Resume | ★ 101 | AI Tools | TypeScript | resume-builder · ats-optimization · job-matching · mcp-server · langchain · langgraph · local-first · privacy · nextjs · typescript · ai-editor · mock-interview · translation · pdf-export · indexeddb · bring-your-own-key | MIT-licensed AI resume workbench (Next.js/TypeScript) that edits resumes in place with multi-model AI, scores and tailors them to job descriptions, and ships a schema-aware MCP server so Claude Code, Cursor or Windsurf can read and patch resumes safely. Self-hosted mode is browser-only with data in IndexedDB; bring you | View Project → |
| metrik | ★ 100 | AI Tools | Rust | AI coding agents · token usage · quota tracking · usage analytics · desktop app · local-first · privacy · reports · multi-agent | A cross-platform desktop app that tracks quotas, reset times, and token usage across locally installed AI coding agents. It parses local logs and official quota endpoints, with reports and no cloud service. | View Project → |
| Librechat-Mobile | ★ 100 | AI Tools | Kotlin | AI chat · mobile app · LibreChat · Android · iOS · self-hosted · multi-account · MCP · streaming · multimodal | Native Android and iOS client for self-hosted LibreChat servers. Supports streaming chats, multiple accounts, model selection, attachments, artifacts, voice features, agents, and MCP configuration. | View Project → |
| blink | ★ 79 | AI Tools | TypeScript | code search · natural language · codebase navigation · CLI · developer tools · file discovery · TypeScript · Jev | Blink is a TypeScript CLI that uses Jev to search codebases with natural-language queries. It ranks likely files by walking directory paths and can run multiple walkers to explore recursively. | View Project → |
| jev-curate | ★ 74 | AI Tools | Rust | dataset curation · synthetic data · pretraining · fine-tuning · LLM evaluation · reasoning data · Parquet · JSONL · streaming · CLI · Rust · Python | A Rust CLI and Python-bindable pipeline that streams Parquet or JSONL datasets and uses TypeSafe AI's Jev model to score, filter, and curate examples. Includes presets for reasoning quality, anti-sycophancy, and code correctness. | View Project → |
| semdecide | ★ 73 | AI Tools | Python | cli · unix-pipelines · semantic-judgment · llm-classification · jsonl · ci · agent-safety · guardrails · exit-codes · python · shell-integration · typed-decisions · confidence-threshold · prompt-free | Python CLI that brings typed LLM judgments to Unix pipelines and CI: is/choose/score/filter commands return predicates, routes, scores, and stable exit codes from TypeSafe AI Jev, plus a guard recipe for agent action safety. | View Project → |
| curlora | ★ 53 | AI Tools | Jupyter Notebook | lora · peft · fine-tuning · continual-learning · catastrophic-forgetting · cur-decomposition · llm · parameter-efficient · matrix-decomposition · research-code · sft · quantization-planned · transformers · pytorch | Research code for CURLoRA, a LoRA variant that uses CUR matrix decomposition to fine-tune LLMs with fewer trainable parameters and less catastrophic forgetting. Includes PyTorch/Transformers implementations and notebooks reproducing experiments on Mistral 7B and GPT2-Large. | View Project → |
| brain-portal | ★ 19 | AI Tools | TypeScript | ai · knowledge-base · notes · rag · embeddings · mcp · self-hosted · vector-search · agents · nextjs · turso · sqlite-fts5 · entity-graph · openrouter · personal-crm · voice-capture | Self-hosted AI-native knowledge base for notes, tasks, projects and contacts. Every item is embedded and FTS5-indexed for semantic Q&A, and a 38-tool MCP server lets Claude Code, Cursor or other agents read and capture into your workspace. | View Project → |
| collabosm | ★ 15 | AI Tools | Python | LLM inference · model serving · OpenAI-compatible API · Google Colab · A100 · ExLlamaV3 · KV cache · Qwen | Scripts to deploy Qwen3.8-Flash-Next on a high-RAM Colab A100 and expose it through an authenticated OpenAI-compatible API. Includes runtime and model setup, session recovery, performance measurements, and KV-cache management. | View Project → |
| typesafe-ai-playground | ★ 5 | AI Tools | Rust | rust · cli · llm-api · phi-detection · code-review · line-importance · tui · text-classification · tree-sitter · syntax-highlighting · streaming · healthcare · interactive · scoring · terminal-ui | Rust CLI playground for TypeSafe's Jev (System One) model API. Commands score text for PHI, rate code-comment accuracy and usefulness, build line-importance heat maps, analyze typed tone live, and classify businesses or occupations. | View Project → |