Custom agents

Tours and activities search via Amadeus — agents find bookable experiences by location, paid per query in USDC on Base or Solana via x402

POI concierge over OpenStreetMap Overpass with optional Yelp enrichment — restaurants, cafes, anything, per query, paid in USDC on Base or Solana via x402

Community, releases, and Building Kavor: a local-first visual workspace for orchestrating coding agents and durable engineering context.

Admin dashboard and CLI for Statewave — manage episodes, memories, and subjects.

Local CLI that sends a computer-use agent's Chrome through global residential and mobile IPs. npx aluvia-cli setup

Terminal bridge that drives browser AI chats and exposes sandboxed local repo tools through MCP.

国家开放教育平台 / 国开大，刷课+AI刷题智能agents油猴脚本

Arbitration for failed real-world tasks — signed case records with evidence hashes, status snapshots, and signed rulings

The skill.md toolkit: generate and validate agent-discoverable skill files (the agentres.dev pattern) from OpenAPI specs. Defines the SKILL-MD format used across the x402 Suite.

The 'code sent to your email' step, solved for agents — paid relay mailboxes, retrieved OTP codes returned in-response

The cancel-worse-book-better move, automated — scans real inventory (Amadeus/RIDB) against your current booking, returns an actionable improvement report

Open-source social media scheduling for 30+ platforms, with a public API, SDK and MCP server.

AI-native parametric CAD where agents propose typed patches and humans review geometry like code. Rust + OpenCASCADE.

Research, Evolve, and Document for AI-assisted engineering

Native macOS terminal for running many coding agents in parallel. Swift 6 + AppKit, SwiftTerm engine, agent state from Claude Code hooks instead of output parsing. Workspaces, split panes, triage dashboard. ~90 MB resident, 2.4us added input latency. brew install --cask essedev/relay/relay

A Telegram control plane for local AI workflows with durable sessions, scheduling, recovery, and permission-aware automation.

A model-agnostic guide for creating concise, project-specific AGENTS.md and CLAUDE.md instructions.

AI-native PCB design tool. The agent designs the board, the human watches and steers. Schematic → routing → fab-ready zip in one loop.

Live marketplace intelligence over eBay Browse — listings search and is-this-a-deal verdicts

Event search and on-sale checks over Ticketmaster Discovery — agents find real events and ticket windows, paid in USDC on Base or Solana via x402

A queryable code symbol graph for multi-language monorepos — find callers, impact, dead code, and cross-file inconsistencies, every edge confidence-scored.

Provenance and decay for AI agent memory. Audits CLAUDE.md, AGENTS.md, .cursorrules, GEMINI.md and 18+ agent memory files against the live repo: typed claims, cited commits, half-life decay, token meter, approve-only fixes. Works with any agent.

AI Agent 技能集合：书籍蒸馏技能库（24本书，85个skills；cangjie-skill流水线试点）+ 项目理解/信息改写/开发UX/小红书发布 + Hermes 运维。兼容 Codex/Claude Code/Hermes

Open MVP for a certified personal-agent workforce marketplace

Transcribe audio from your coding agent — the Agent Skill, CLI, and SDK for OpenTranscription.

HERALD is an open-source CLI that emits the standard discovery files agents need to read your site and discover its agent-interaction capabilitie. One config object drives all of it. Each file is an independent open standard; pick the layers you want and HERALD generates only those.

Generative UI specification for native mobile: a declarative JSON language in which AI agents write a whole feature, its data, logic and navigation included, and the iOS and Android runtimes run it natively on SwiftUI and Jetpack Compose, offline.

Signed, hash-addressed decision records for AI agents, with reasons you can re-run

a collection of custom Claude Code skills for productivity and project management

same cut, every project — repository procedures installed as skills into Claude Code, Codex, and Antigravity, with the repository state converged to match.

Facet compares what a brand says—its positioning—with what its product actually delivers. It traces the promise across marketing, default UX, policies, and observed outcomes, then returns a source-backed verdict. | Positioning Audit

Fast System-1 Decision, Arbitration & Safety Engine for Autonomous AI Agents (Powered by TypeSafe Jev)

Vibe code from one app: run agents in parallel worktrees, watch their diffs live, browse, test and commit. A 14 MB git client with a file explorer and terminal built in.

Documentation-first development for AI coding agents — 10 skills: persistent memory, zero-ambiguity specs, append-only decisions, and an independent verification gate. The harness layer for long-running, loop-driven development.

Make a domain's transactional valid operations the AI agent's action space — invariant-preserving, approval-gated, revertible, and verifiable.

An MCP server for discovering, ranking, and retrieving agent skills while minimizing context usage.

Manage a fleet of herdr machines over Tailscale: dashboard plugin, auto-discovery, capacity-aware agent dispatch, diskless workspaces

Go/no-go weather decisions for plans — hourly forecasts and verdicts from Open-Meteo and NWS, keyless and live, paid with x402: USDC on Base or Solana.

herdr plugin: auto-name panes & tabs after each agent's live topic (Claude Code, Codex, …) instead of 1, 2, 3.

Distribution amplifier for solo AI creators. Submit your AI project once — agents auto-write 10+ platform-native posts in ~10 seconds. Next.js 16 · Supabase · Claude.

Turn YouTube, websites, and arXiv papers into a structured, reusable corpus of insights, syntheses, and reports.

Least-privilege governance for ERPNext — the credential floor + a governed agent broker, MCP · A2A · API, one spine. No debit without a credit: PLAN · CONSENT · PROVE · UNDO.

Spec-driven development kit that orchestrates a team of AI agents (Copilot, OpenCode, Claude Code) through the full change lifecycle.

Blind-test products through real desktop and browser UI with isolated AI personas.

AI-powered Dropshipping Agent that automates product research, competitor analysis, and profitability prediction using Python, web scraping, and LLMs.

Screen Beaver is a Windows screen recording and product demo studio that turns raw recordings into polished demos automatically. Record your screen or Android device, add smooth zooms, cursor effects, captions, and more, or use MCP to create demos hands-free through your coding agent.

SENTINEL: an authority-attenuating immune architecture for AI agents. Detect and contain compromised agents without giving the defenders unrestricted authority.

Turn one ZCode conversation into a restartable multi-agent workflow with persistent memory, isolated coding agents, real-result checks, and human-approved merge.

Google Cloud integrations for Twilio Agent Connect — connect Google Cloud agent runtimes to Twilio's communication channels.

The context layer for AI agents — open, structured, searchable knowledge protocol (Domain → Concept → Link). Skills-first, no MCP server needed.

Anti-slop AI UI skills for web & mobile — design direction, design systems, motion, responsive UI, accessibility, assets, and visual QA.

Intlayer Skills are a set of specialized instructions and workflows for AI agents to help them interact with Intlayer projects.

Transit trip planning and live delay snapshots from open GTFS/GTFS-RT feeds, priced per query

Receipts and accounting for agent spending — enriched on-chain receipt lookups and ledger exports

Pay-per-query flight search over the Amadeus API — offers, live pricing, and fare-drop checks for agents

This is a free and open-source cryptocurrency trading bot written in Python.

Local-first cognitive runtime for AI agents: bounded context, persistent memory, tool/model routing and verified execution.

Route Claude Code, Cursor & Codex through the TokenSaver control plane — Zero Trust governance, MCP, Flux IA. Free OSS CLI.

Self-hosted, open-source platform for running autonomous AI agents 24/7 with sandboxed Docker containers, persistent memory & workspaces, task scheduling, workflows, and live streaming. Pluggable brains: Claude Code, Codex, opencode. Your server, your data.

Find and scrub secrets from AI coding agent session logs and configs: Claude Code, Codex, Gemini CLI, aider, Cursor and more

MCP server for Ticketmaster - let AI agents buy event tickets

Runs BOTH controls (nop + oracle) across EVERY task in a coding-agent benchmark, classifies the outcome, captures evidence, and lints leaks statically. One binary, zero LLM calls.

Hunt as a pod — Coordinate autonomous AI agents that work together in habitats. Orcys form a pod, create missions, claim tasks, and hunt together.

🎥 edit videos like a pro

Herds the yaks so your agent stops shaving them. A strict SSOT governance harness for agentic coding.

OBJECTION! A skill and plugin for AI coding agents (Claude Code, Codex, Cursor, Gemini): accusers, a defender and a judge review the PR, and a gate blocks it until APPROVED.

A portable Agent Skill for designing, refactoring, and reviewing reusable skills.

The agent-native application SDK. Define schemas, views, procedures and triggers in YAML. Mantle compiles them into native SQL and typed TypeScript, with optional REST APIs, an MCP server for AI agents, Better Auth, llms.txt-ready web pages and an admin console. Runs on Cloudflare Workers and Bun with SQLite, D1 or PostgreSQL.

A demo of AI-hosted live shopping, built on LemonSlice. Live video rooms with AI avatar hosts that pitch products, answer questions, and close the sale.

HarnessTemplates | Harness Engineering 中文模板, 旨在释放双手, AI全自动

Auto-generate Model Context Protocol servers from ABP Framework apps. In-process, permission-aware, multi-tenant.

Persistent work graphs and independently checked quality gates for autonomous Claude Code builds.

An AI creative workflow engine — a small vocabulary of primitives, the model plans the work. First product in the Lumeri family: Lumeri Video.

Agent skill that audits code and plans with connected GitHub context and a strong ChatGPT model using Effort Pro through Chrome.

Grid carbon intensity now and best-window scheduling — keyless live data for green agents, paid with x402: USDC on Base or Solana.

Class and session booking server — fitness studios and workshop hosts sell seats per-class via x402

EnigmAgent — 6-layer encrypted vault for AI agents (AES-256 + ChaCha20 + Blowfish + Fernet + 3DES + RSA). P2P sync via Gun.js. Zero cloud. Part of P2PCLAW.

Open, independent benchmark on web search APIs for AI agents - factual lookup on 300 company news questions, ranked by cost per 1,000 correct answers. TinyFish, Parallel, Perplexity, Linkup, Firecrawl, Brave, You, Exa, Tavily, Google SERP.

Local mission control for AI coding work: durable missions, evidence gates, and guarded completion.

适用于 Claude Code 和 Codex 的求职 Skill：查找岗位、判断是否值得投递、制作简历与申请材料、检验申请材料并模拟面试。不编造技能或业绩数字，不预测录用概率；检查简历是否便于招聘管理系统（ATS）读取，并从招聘人员和用人经理视角分别独立评审。

Hybrid SD + LLM agent-based simulation of engineering feedback degradation under different product management regimes. Tests Schelling focal points (committed budgets, error budgets) as structural interventions. Built in AnyLogic 8.9.9.

Autonomous web-app QA skill for Claude Code: Playwright E2E, visual regression, axe-core a11y. LLM-generated self-healing specs, zero-token test runs.

Pure Jev that can "type" and drive towards task completion.

Independently-verified multi-agent delivery for any task. A Caster picks the crew, and every workstream passes a mandatory honesty gate. Runs on Claude Code, Codex, Gemini, and CodeWhale.

a multi-utility IRC (and Twitch) chatbot

AI-powered platform that helps Destination Management Companies (DMCs) discover, source, and coordinate tourism suppliers and experience creators, then build tailored tourism itineraries and manage their operations.

Harness & recipes for efficient retrieval LLM agents

Run and Fine-tune GGUF models, with LoRA or full weights, from SFT to agentic GRPO

Have an LLM (LangGraph agent) control QuantLib, the cache, your databases and any plug-in tools. Includes evaluator framework and an Excel/LibreOffice API in https://github.com/FulgentMcGuffin/cqfi-xl and makes use of https://github.com/FulgentMcGuffin/mcp_data

Speaker-aware MCP server for searchable podcast transcripts.

The trust baseline for multi-agent memory — event-sourced, governed, federated. 事件溯源 · 治理管线 · 联邦同步

面向高校课程与科研的 AI 智能体平台（LangGraph）：课程助教 / 题库练习 / 学情分析 / 科研助手，知识检索由 omnigraph 提供

Curated list and benchmark leaderboard of ultra-lightweight, zero-dependency C/C++/Rust/Assembly engines for local LLMs, edge inference, vector search, and AI agents (2026-2027)

OSINT cross-event analysis: news + open sources → a scored knowledge graph. LLM entity/event extraction, TMFG themes, cross-story bridges, Louvain storylines, and evidence-backed ICD-203 claim verdicts.

📱 Phone control system - Remote device management & automation

Source-grounded atomic learning, RAG, research reading, exam preparation, and persistent personalized study for Codex.

A JSON-encoded, SQL-flavoured predicate language described by a single JSON Schema — $ref it from an OpenAPI document, or inline it into an MCP tool's inputSchema. One filter grammar for every search endpoint in an API.

Capture browser API traffic via CDP and turn any site into a ready-to-run FastMCP server, with no official API required.

An agent with an executive function. Metis uses a core reasoning matrix to decide how to think, dynamically creating its own perceptual faculties for each problem. This is a new architecture for building autonomous agents whose strategy and identity emerge directly from a symbolic protocol.

AI-assisted host provenance and incident-response workspace