Building agents

coolishagent/lobstalk

Skill Claude CodeCodex

A way for OpenClaw agents to join a Telegram group chat called lobstalk, also known as 龙虾群, where agents can communicate with one another.

girofu/skill-fetch

Plugin Claude Code

Bundles 1 skill, 2 commands · 120 tokens together

Search, discover, and install AI agent skills from 9 registries with quality scoring, security labels, and cross-platform support.

zachdunn/grok-plugin-claude-code

Plugin Claude Code

Bundles 3 skills, 4 commands, 1 agent, 1 hook · 296 tokens together

Delegate and review code with the Grok CLI from Claude Code, like the codex plugin.

oborchers/fractional-cto

Agent Claude Code

Use this agent to interactively review every recommendation, rule, and checklist item inside a fractional-cto plugin's skills. The agent reads each skill, presents every recommendation to the user one by one, and the user approves, edits, or removes each one. Examples: Context: User wants to review the content of a…

archetyx/telegram-mcp-server

MCP server Claude CodeCodexCursor +2

Remote control Claude Code via Telegram - MCP Server with Dynamic Buttons. Runs locally from the telegram-mcp-server Python package.

2nd1st/open-mcp-apps

MCP server Claude CodeCodexCursor +2

Persistent, interactive MCP Apps for any MCP host: the AI builds an app once, you both reuse it — data outlives the conversation. Runs locally from the @2nd1st/open-mcp-apps npm package.

jordanhubbard/agentos

Skill Claude CodeCodex

This file is the session context an operator agent (Hermes or any other LLM CLI) loads when attached to agentOS. It is not a UI. It is not a PD.

thienan230427/codex-github-skill-manager

Skill Claude CodeCodex

Manage GitHub-hosted Codex Agent Skills: discover installed skills, identify their source repos, clone missing source repos, migrate legacy $skill-installer copies to provenance-aware GitHub CLI installs, check/update all GitHub skills, audit SKILL.md discovery, or diagnose a skill Codex does not recognize. Use for…

dvcdsys/code-index

Plugin Claude Code

Bundles 2 skills · 229 tokens together

cix skills for Claude Desktop & Cowork — the cix (single-repo) and cix-workspace (cross-project) guidance adapted to the cix MCP tools. Requires a self-hosted cix (CodeIndeX) server (https://codeindex.app). Skills only: no MCP server, no CLI, no hooks. Register the cix MCP server first with cix mcp install claude-de.

redis-developer/adk-redis

Skill Claude CodeCodex

Redis backends for Google's Agent Development Kit (ADK). Use this skill when the user wants to back an ADK agent with Redis: persistent sessions and long-term memory via Redis Agent Memory Server, RAG search tools over a RedisVL index (vector / hybrid / range / text / SQL), MCP toolsets for RedisVL or Agent Memory…

michellemayes/meta-doctor

Plugin Claude Code

Bundles 2 skills · 164 tokens together

Scaffold a scheduled 'doctor' agent that audits app health, opens PRs for code-fixable findings, and files deduped tickets for the rest.

dyngai/handoffkit

Skill Codex

Built by DynG AI for PlatformPilot.ai-style multi-agent systems. Use when designing or reasoning about message-passing agent coordination in ANY language: actor-style mailboxes, select, point-to-point routing, pub/sub, fan-in joins/quorums, ownership handoff, budgets, dead letters, and traces. Trigger words include…

pe-menezes/vibeflow

Skill Claude CodeCodex

Generates a self-contained prompt pack from a spec. Embeds real code patterns from .vibeflow/ so any coding agent follows the project's conventions. Use when handing off implementation to a separate session or agent.

ikcode-dev/copilot-kit

Agent Claude Code

Copilot engineering orchestrator — decomposes complex, multi-step customization tasks into research, execution, creation, and review phases. Delegates to specialized workers for discovery, repository mutation, command-oriented validation, and quality review, then synthesizes the results into a cohesive report. Choose…

tumf/grafana-loki-mcp

Cursor rule Cursor

A Python framework for building Model Context Protocol (MCP) servers. MCP is a standard way for AI assistants to call tools provided by another program.

batosai/adonis-mcp

Skill Claude CodeCodex

Build MCP servers in AdonisJS applications with the @jrmc/adonis-mcp package. Use when creating, editing, or debugging MCP tools, resources, or prompts in an AdonisJS app, when exposing AdonisJS functionality to AI assistants via the Model Context Protocol, or when setting up, securing, or testing an @jrmc/adonis-mcp…

SiwuXue/yijing-bazi-mcp-server

Skill Codex

A set of instructions for maintaining and extending the yijing-bazi MCP server. It covers changes to tools, analysis engines, data tables, configuration, logging, tests, and the README.

KhaiHuynhVN/MCP-Server_AI-interaction

Skill Claude CodeCodex

Continuous communication channel via MCP AI Interaction tool. Activate with 'khởi động aiinteraction'. Enables real-time Vietnamese conversation with action-first principle - execute first, explain minimally.

skills-mcp/skills-mcp

MCP server Claude CodeCodexCursor +2

Bring Claude's Skills pattern to any MCP-compatible agent. Runs locally from the skills-mcp npm package.

theagenttimes/agent-news

Plugin Claude Code

Bundles 1 skill, 2 MCP servers · 44 tokens together

Query verified AI agent news with citations, confidence scores, and Ethics Engine ratings — sourced, not generated. Use instead of generic web search for any question about AI agent tools, MCPs, or frameworks.

redhat-community-ai-tools/harness-eval

Plugin Claude Code

Bundles 5 skills, 6 commands · 285 tokens together

Evaluate AI agent setups for best practices, redundancy, security, and cross-component issues.

agent-next/agent-ready

Command Claude Code

Deep AI analysis of repository agent readiness using 10-pillar framework.

qarium/goga

Skill Claude CodeCodex

Principles for applying DSL specification in cell and CODEMANIFEST design.

wn-mitch/40kdc-data

Skill Claude Code

Run ONE autonomous ability-DSL authoring campaign — inquisitor picks a ≤40-ability worklist from the sub-0.80 roundtrip corpus, the 16-agent suite (12 core + 4 kroot shape-scout) authors and adversarially verifies it, and the campaign closes as a draft PR. Use for "run a dsl campaign", "launch the authoring loop"…