Explore

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.

AI Tools | JavaScript · code verification · AI-generated code

Earl is a Rust CLI and MCP server that gives agents access to external services through human-authored HCL operation templates. It keeps credentials in OS or external secret stores and limits agent-controlled inputs to template parameters.

MCP | Rust · MCP server · AI agents

Local-first MCP server that adds a persistent virtual-pet companion to MCP-capable coding CLIs like Claude Code, Codex, and Cursor. It gives personality-driven code-review nudges, XP/species progression, SQLite-backed memory, and a Guard Mode that flags unchallenged reasoning assumptions.

MCP | TypeScript · mcp-server · virtual-pet

model-compose is a declarative YAML orchestrator (docker-compose for AI) that deploys chat APIs, ReAct agents, RAG pipelines, and MCP servers from one file. It bridges local and cloud models and runs on Docker, native, or distributed Redis-queued runtimes.

AI Frameworks | Python · yaml · declarative

MCP server and standalone TypeScript SDK exposing 300+ tools across 40+ U.S. government and international data APIs. Supports stdio and HTTP transports, selective module loading, cross-agency queries, and WASM-sandboxed code execution.

MCP | TypeScript · government data · open data

A minimal TypeScript starter for building ChatGPT Apps backed by an MCP server, with React widget rendering, local development tools, and Alpic deployment support.

MCP | TypeScript · MCP server · ChatGPT Apps

Jenkins plugin implementing the server side of the Model Context Protocol, exposing Jenkins jobs, builds, and logs as MCP tools and resources over SSE, Streamable HTTP, or stateless transports. Lets LLM agents and IDEs drive Jenkins with the same credentials as the instance.

MCP | Java · mcp-server · jenkins

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.

AI Tools | PHP · laravel · cms

An Android AI assistant and agent with on-device GGUF inference, local document RAG, memory, and built-in tools. It can also connect to hosted or self-hosted model providers and run scheduled tasks.

AI Agents | TypeScript · mobile AI agent · local-first

Malloy Publisher is an open-source analytics engine that serves Malloy semantic data models over MCP and REST, so AI agents, apps and BI tools query governed models instead of writing raw SQL. It bundles agent skills, dashboards, materialization, and row-level access control.

MCP | TypeScript · mcp-server · semantic-modeling

An MCP server that lets AI assistants manage WordPress sites through 71 tools for content, media, users, SEO, and site operations. Supports multiple sites and integrates with clients such as Claude Desktop.

MCP | JavaScript · WordPress · CMS

An MCP server and CLI that captures webpages and returns screenshots sized and tiled for AI vision workflows. Supports full-page capture, CSS selector screenshots, screencasts, and saving images to disk.

MCP | TypeScript · MCP server · web screenshots

An open registry of affiliate programs with structured program data, a public API, CLI, SDK, and MCP server. AI agents can search programs and retrieve details such as commissions, cookie duration, and payout terms.

MCP | TypeScript · MCP server · affiliate programs

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.

AI Tools | Python · LLM router · API proxy

An MCP server that lets Claude and other MCP clients inspect, edit, generate, and validate Final Cut Pro timelines through FCPXML. Includes transcript-based editing, media analysis, timeline QC, and optional import into a running Final Cut Pro.

MCP | Python · Final Cut Pro · FCPXML

GitHits provides a CLI and MCP server that lets coding agents search indexed open-source code, read source and documentation, find examples, and inspect package health. It integrates with multiple coding tools through local or hosted MCP setup.

MCP | TypeScript · code search · code navigation

A curated, health-tracked catalog of MCP servers for geospatial work, covering mapping, geocoding, routing, spatial databases, earth observation, and GIS. Includes a Claude Code skill for searching the catalog and adding server listings.

MCP | Python · geospatial · GIS

TypeScript CLI, HTTP API, MCP server (22 tools) and agent skills giving AI agents a shared layer to search live supermarket catalogues, compare prices, build baskets and preview checkout across 10 providers in 7 countries. Batch primitives and optional Open Food Facts enrichment included.

MCP | TypeScript · grocery · supermarket

A local Rust daemon that stores and retrieves semantic memories for AI assistants, with a CLI, browser UI, and MCP interface. Uses LanceDB and Qwen3 embeddings, and integrates with Claude Code, Codex, Cursor, Zed, and OpenClaw.

RAG | Rust · semantic memory · local-first

A catalog of go-to-market tools and workflows, exposed through a read-only MCP server. Agents can search entries, fetch their Markdown instructions, and use workflow prompts to carry out documented growth tasks.

MCP | TypeScript · MCP server · go-to-market

MicroCore is a minimalist Python library of LLM and vector-DB adapters that makes providers switchable via config while keeping app code unchanged. It includes prompt templating, streaming, embeddings search (Chroma/Qdrant) and LLM-agnostic MCP tool integration.

AI Frameworks | Python · llm · mcp

ACC lets independent AI coding sessions (Claude Code, Codex, Gemini CLI, Grok, Kimi) message each other, leave handoffs, and request reviews. Runs locally with a CLI, hooks, and an MCP server; no hosted service or lead agent required.

AI Agents | JavaScript · agent-communication · multi-agent

A self-hosted, n8n-orchestrated ChatOps system that uses tiered AI agents to triage infrastructure and security alerts, investigate causes, and propose fixes. It combines graph-based impact prediction, monitoring, and human approval gates for operational changes.

AI Agents | Python · agentic-ai · chatops

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.

AI Tools | Python · video-generation · video-editing

A step-by-step Java workshop for building AI applications with Quarkus and LangChain4j, covering single AI services and agentic orchestration. Each lesson has a runnable project state.

AI Frameworks | Java · workshop · Quarkus

A daily updated directory ranking GitHub repositories for academic research and writing agent skills. Includes searchable rankings, agent compatibility labels, trend history, and JSON, CSV, and RSS feeds.

Skills | JavaScript · academic research · agent skills

A local or HTTP MCP server that lets AI clients query and analyze Langfuse traces, sessions, exceptions, metrics, prompts, and datasets. Includes a Langfuse agent skill and supports Claude Code, Codex, Cursor, and other MCP clients.

MCP | Python · Langfuse · observability

pi-ling is an Electron desktop workbench for coding agents that runs Native, DSH and Codex runtimes in one session. Every tool call, file edit and model output lands in a SQLite event log, with effect-based approval policies, file-based skills, subagents and a coding eval toolchain.

AI Coding | TypeScript · coding-agent · agent-runtime

Local-first hybrid search engine for Obsidian vaults, combining BM25, fuzzy title/alias matching and sqlite-vec semantic embeddings fused with RRF. Ships a CLI, an Obsidian plugin, and an MCP server so agents can search, read and traverse notes as tool calls.

RAG | TypeScript · obsidian · hybrid-search

A local-first platform that dispatches software tasks from a Kanban board to role-based agents running through Claude Code, Cursor, Antigravity, OpenCode, or model APIs. Coordinates implementation, review, QA, and release workflows on your machine.

AI Coding | Go · coding-agents · multi-agent

A curated directory of research, standards, benchmarks, and tools for securing AI agents and their skills. Covers attacks such as prompt injection and tool poisoning, plus defenses including access control, sandboxing, validation, and red teaming.

AI Agents | agent security · LLM security

OpenChatX is a local agent runtime that connects ChatGPT to computer-control tools, files, shell commands, custom capabilities, and other MCP servers through one MCP connection. Includes skills, rules, projects, workflows, and configurable subagents.

AI Agents | TypeScript · local agent runtime · computer use

Self-hosted MCP memory server giving AI agents durable, scoped memory plus hybrid dense+SPARSE+Neo4j graph retrieval with citations. Ships Docker stack, REST/OpenAI-compatible APIs, connectors, and eval harnesses.

MCP | Python · agent-memory · rag

An MCP server that lets AI assistants generate, validate, and render UML and other diagrams from natural-language requests or source text. Supports about 37 Kroki-backed diagram types, including PlantUML, Mermaid, D2, BPMN, and C4.

MCP | Python · UML · diagram generation

Kiln is a local MCP server, CLI and TypeScript library that lets a coding agent write JavaScript programs to build game-ready GLB models, render views, review results and revise. Includes authoring, editing and QA skills for harnesses like Claude Code and Codex.

MCP | TypeScript · text-to-3d · 3d

Go MCP server bridging MCP clients to DAP-compatible debuggers so AI agents can start sessions, set breakpoints, step, inspect stack/variables, attach to processes, and debug core dumps. Exposes 13 tools over stdio; includes agent skills and prompts.

MCP | Go · mcp-server · dap

Superpower is a browser extension plus Qt desktop app that brings MCP tool execution into web AI chats like ChatGPT, Gemini and Grok. It detects structured tool calls, runs local/remote MCP servers with review-first guarded workflows, and adds page/YouTube summarization and local knowledge capture.

MCP | TypeScript · mcp-client · browser-extension

CHAP is an open protocol and runtime for auditable human-agent work: approvals, overrides, handoffs and escalations captured as hash-linked envelopes. It ships MCP and A2A servers plus Python/TypeScript reference coordinators with SQLite persistence and conformance vectors.

AI Agents | HTML · human-in-the-loop · audit-trail

Official Harness MCP server exposing the Harness.io platform to AI agents via 11 consolidated tools and 251 resource types across CI/CD, GitOps, feature flags, cloud cost and security. Supports stdio, HTTP, OAuth and multi-user deployments with 35 prompt templates.

MCP | TypeScript · mcp-server · harness

An experimental, local-first coding and office agent with TypeScript and Python runtimes, desktop and CLI clients, tool permissions, session tracing, MCP and skills support, and multi-agent workflows. It can work with code, documents, spreadsheets, slides, PDFs, and browser tasks.

AI Agents | Python · AI agent · coding agent

A self-hosted, multi-tenant platform for building persistent AI agents with tools, skills, sub-agents, memory, MCP connections, and scheduled triggers. Supports hosted providers and local models, with sandboxed shell and filesystem access.

AI Agents | TypeScript · self-hosted · multi-tenant

A self-hostable MCP hub for discovering, configuring, and aggregating MCP servers for AI clients. It also provides document storage, semantic search and RAG, persistent AI memory, workspaces, and activity logs.

MCP | TypeScript · MCP client · MCP server

Single Go binary wrapping llama.cpp into a self-hosted AI assistant: chat UI, persistent Markdown memory, terminal/file/web/browser tools, MCP servers, encryption at rest and encrypted remote access. Handles GPU detection, model presets and an OpenAI-compatible endpoint.

AI Agents | Go · local-ai · self-hosted

An MCP server that gives AI clients UI5-specific development tools, including project scaffolding, API reference lookup, guidelines, project inspection, manifest validation, and linting.

MCP | TypeScript · MCP server · UI5

A suite of Claude Code plugins for multi-agent software delivery, covering project setup, implementation planning and coding, automated PR reviews, quality judging, and organizational learning. Includes workflow gates grounded in the codebase.

AI Coding | Python · Claude Code · multi-agent

A Rust MCP server that gives coding agents repository code maps, document RAG, shared memory, agent communication, web crawling, and git history tools. Supports 300+ languages and integrates with more than 10 coding-agent harnesses.

MCP | Rust · MCP server · coding agents

A Python API bridge that connects Ollama with multiple MCP servers and makes their tools available during chat requests. Supports stdio, HTTP, and SSE transports, tool filtering, streaming, and transparent proxying of other Ollama endpoints.

MCP | Python · Ollama · tool calling

MCP server exposing 333 wire-format-verified tools for the pfSense REST API, letting Claude Desktop/Code manage firewall rules, VPN, DNS, DHCP and diagnostics in natural language with a confirmation gate, config backup and rollback on destructive changes.

MCP | Python · mcp-server · pfsense

Chrome side-panel extension that converts a page's WebMCP tool schemas into typed Jev questions to predict tool calls, arguments, and confidence from user input, with per-site permissions, manifest screening, and confirmation gates for state-changing tools.

MCP | JavaScript · chrome-extension · webmcp

Chimp is a Scala 3 SDK from SoftwareMill for building MCP servers and clients with type-safe, boilerplate-less APIs on Tapir and sttp. Supports streamable HTTP and stdio transports plus streaming integrations for Ox, ZIO and Pekko.

MCP | Scala · scala3 · sdk

A DeepSeek Harness plugin that gives agents cross-session memory using layered conversation summaries, persistent event and graph storage, and evidence-linked retrieval. Older context becomes more concise while original records remain available for verification.

AI Agents | TypeScript · agent memory · cross-session memory

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.

AI Tools | HTML · AI security · LLM security

Rust CLI and MCP server for checking Forma benefits and submitting claims. It can use an LLM to extract claim details from receipt images or PDFs, with confirmation before submission.

MCP | Rust · MCP server · Forma

A Python framework for building domain-specific ReAct agents with configurable LLMs, tools, memory, skills, RAG, and MCP integration. Includes a FastAPI service with streaming output and example task workflows.

AI Agents | Python · AI agent · ReAct

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.

AI Tools | Python · knowledge management · AI summarization

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

AI Tools | TypeScript · resume-builder · ats-optimization

Rust/Tauri desktop AI work system with multi-persona agents, long-term compressed memory, MCP and skill management, reversible tool execution with AI review, and remote IM bridges. Local-first, no intermediary servers; aimed at long-running tasks rather than chat.

AI Agents | Rust · desktop-ai-assistant · multi-agent

A TypeScript project-memory tool for Claude Code, Cursor, and Codex. It records decisions, constraints, bugs, and failed approaches, retrieves relevant notes for later prompts, and can check tool calls against saved rules.

AI Coding | TypeScript · coding-agent-memory · Claude Code

A self-hostable AI workstation for multi-agent tasks, cited knowledge search, skills, and MCP-connected tools. Supports multiple model providers, sandboxed execution, and web and desktop clients.

AI Agents | TypeScript · multi-agent · self-hosted

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.

AI Tools | Rust · AI coding agents · token usage

A Python MCP server that connects AI clients to Basecamp 3 through OAuth. It exposes tools for projects, search, todos, messages, card tables, documents, uploads, and other Basecamp resources.

MCP | Python · Basecamp 3 · project management

Compiler-backed code intelligence infrastructure for AI coding agents. It builds a tree-sitter index with resolved symbols, call relationships, and data flow, then exposes structured query, edit, test, and rollback tools, including MCP compatibility.

AI Coding | Python · AI coding agent · code intelligence

An MCP server that lets AI assistants read and control DaVinci Resolve projects, timelines, media, color, and rendering. Includes local transcription, vocal isolation, and background removal using open-source AI tools.

MCP | Python · DaVinci Resolve · video editing

An MCP server and CLI for controlling Windows applications through the Windows UI Automation API. Agents can inspect controls and click, type, select, and read by element identity, with screenshot and coordinate input as a fallback.

MCP | C# · Windows automation · UI Automation

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.

AI Tools | Kotlin · AI chat · mobile app

Self-hosted AI workspace for chatting with models, building agents, deep research, and code generation. Connect provider API keys directly; agents can use bundled skills and tools, including Gmail and Calendar integrations.

AI Agents | TypeScript · self-hosted · BYOK

A local MCP server that lets AI clients control Logic Pro through structured tools and inspect live project state. Mutations use explicit targets, confirmation checks, and readback to report confirmed, uncertain, or failed outcomes.

MCP | Swift · Logic Pro · music production

A community-driven website where visitors vote on feature ideas and Claude Code implements the winning proposal every 30 minutes. The project includes prompt moderation, diff restrictions, build checks, deployment verification, and automatic rollback.

AI Coding | Astro · autonomous coding · AI code generation

Research code and preview for ScienceBuddy, an interactive scientific agent workspace plus a double-recursive self-improvement experiment that alternates harness refinement with SkyRL GRPO model training on frozen scientific tasks.

AI Agents | Python · scientific-agents · interactive-agent

A large hands-on lab of Jupyter notebooks and Python projects covering LangGraph and AutoGen agent systems, MCP servers, production RAG with reranking, mem0 memory, n8n automation, plus fine-tuning and vLLM deployment. Best for developers learning end-to-end agentic AI by example.

AI Agents | Jupyter Notebook · agentic-ai · multi-agent

A Claude Code plugin that judges each large tool result with a calibrated System One model (Jev, or a Haiku adapter) and replaces unneeded blocks with a stub plus a recall key. Hooks, shadow mode, and MCP server included; local cache keeps the hidden text restorable.

AI Coding | Python · claude-code · claude-code-plugin

A TypeScript supervision layer for Claude Code and Codex CLI. Hooks record agent actions and block completion when code changes lack a passing check; deterministic rules can also catch secrets and test removal. Jev-backed judgments are advisory.

AI Coding | TypeScript · coding-agent-guardrails · agent-hooks

Cyber Xiaowan is a public SKILL pack for Codex/Claude Code that acts as a lightweight structured-thinking operating system: it listens, clarifies, audits, architects, or suggests one small next step. Ships an npx installer, progressive-loading reference rules under references/, and an experimental local desktop-pet mod

Skills | JavaScript · skill-pack · claude-code

AMA-Bench is an ICML 2026 evaluation framework for agentic memory: methods build memory from long agent trajectories, retrieve evidence, and answer QA scored by LLM-as-judge. Includes vLLM/API pipelines, cross-judge validation, and a HF leaderboard.

AI Agents | Python · agent-memory · long-horizon

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.

AI Tools | TypeScript · code search · natural language

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.

AI Tools | Rust · dataset curation · synthetic data

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.

AI Tools | Python · cli · unix-pipelines

MemoryArena is a research framework and benchmark for agent memory in interdependent multi-session agentic tasks. It wires pluggable memory backends (long-context, mem0, Letta, Mirix, GraphRAG, MemoRAG, BM25) into task agents and step-based environments for web shopping, travel, search, and formal reasoning.

AI Agents | Python · agent-memory · benchmark

A local stdio MCP server that exposes TypeSafe Jev judgments to Cursor, Codex, and other MCP clients. It supports coding-step routing, prepared-call selection, patch review, evidence checks, content screening, and candidate ranking; the host agent executes tools and edits files.

MCP | TypeScript · MCP server · coding workflow

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.

AI Tools | Jupyter Notebook · lora · peft

Official ICLR 2024 implementation of HOMER, a training-free hierarchical KV-cache merging method that extends pre-trained LLM context limits (e.g. Llama-2) with lower memory. Ships patched LlamaForCausalLM, plus passkey-retrieval and PG19 perplexity scripts.

AI Frameworks | Python · long-context · kv-cache

A browser-based 1v1 quickscope FPS where players fight Jev, an AI opponent driven by the TypeSafe SDK. The server runs the match simulation and uses a deterministic heuristic fallback when the AI service is unavailable.

AI Agents | TypeScript · AI opponent · game agent

A MIT-licensed roster of 200+ markdown AI agent personas (engineering, design, marketing, sales, security, GIS, game-dev) installable as subagents into Claude Code, Cursor, Codex, Gemini CLI and other agentic tools via shell scripts or a desktop app.

AI Agents | Shell · subagents · agent-personas

A Python market-making system that uses Jev for selected market judgments while deterministic code handles state, quote policy, execution, and hard risk vetoes. Includes paper trading, fallback behavior, and decision calibration.

AI Agents | Python · AI trading · market making

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.

AI Tools | TypeScript · ai · knowledge-base

Local Agent skill that turns a character reference image into a short action video, PNG animation frames, and a sprite sheet. Includes Python scripts for image-to-video generation, frame extraction, alignment, and spritesheet building.

Skills | Python · agent-skill · image-to-video

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.

AI Tools | Python · LLM inference · model serving

OpenCode agent framework: installable prompts, agents, subagents and commands for plan-first, approval-based coding workflows, plus an eval harness that runs agent test suites against Claude, GPT-4 and Grok. Built for the OpenCode CLI, not standalone.

AI Coding | TypeScript · opencode · ai-agents

Rules-and-receipts layer for AI agents that spend money: plain-English policy compiles to allow/ask/deny, risky actions hit a human approval inbox, and every outcome produces an Ed25519-signed receipt anchored on Base. Contracts are unaudited; mainnet settlement not yet demonstrated.

AI Agents | JavaScript · agent-governance · agentic-commerce

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.

AI Tools | Rust · cli · llm-api

ServiceEmma is a LangGraph customer-service agent for appointment businesses: it answers FAQs from RAG knowledge, books or cancels appointments, holds slots via Stripe Checkout, and escalates to support tickets. Ships with FastAPI backend, SSE chat widget, Postgres state, voice, and an owner dashboard.

AI Agents | Python · customer-service · langgraph