23,553 tagged Building agents, measured the same way as everything else here.
Browse within: agentic-ai 231LLM 215agent 200agents 152skills 143agentic-workflow 108openai 86agent-orchestration 84antigravity 79agentic-framework 73cli 73ai-skills 71Autonomous Agents 67agent-framework 62
Skill Claude Code
Best practices for building Godot 4.x games through the Godot MCP Gateway. Covers tool selection, workflow patterns, error recovery, and token efficiency for MCP-connected agents.
Skill Claude CodeCodex
Use when the user asks to get an independent ChatGPT web review, continue a GPT-Control conversation, inspect a model run, or generate or iterate on an image through a configured browser driver. Not for facts the working tree or a local test can answer directly.
Skill Codex
Delegate independent coding, review, or research tasks to low-cost DeepSeek Harness agents in isolated Git worktrees, with duration-aware foreground or detached monitoring and Markdown review artifacts. Use when the user explicitly asks for DSH, DeepSeek, external/low-cost subagents, or batch delegation; do not use…
Skill Claude CodeCodex
Delegate coding tasks to user-configured DevSpace subagents.
Skill Claude CodeCodex
A policy for handling messages received through a Matrix impersonation channel. It checks whether the sender is trusted before allowing work and always asks the manager about high-risk actions.
Skill Claude CodeCodex
Supervise a coding task through the dsh-gate MCP tools, including durable reconnect, interaction handling, observation cursors, the five-minute aggregated progress cadence, and strict completion validation. Use when the user asks Codex to delegate or supervise work in a DSH Host. Do not use for ordinary local coding…
Skill Claude CodeCodex
Inspect agent JSONL traces and run deterministic evaluation gates.
Skill Claude CodeCodex
Autonomous MCP server routing, multiplexing, performance benchmarking, and real-time observability suite for AI coding agents.
Skill Claude CodeCodex
Delegate work to 09orche tools, decide when a reusable profile is worth creating, and verify subagent output without a rigid mandatory pipeline. Use when the user asks to delegate a task to another model, create a subagent persona, or set up a multi-model workflow.
Skill Codex
Prepare, save, and transfer durable coding-agent context using OpenHandoff when a task spans agents, sessions, or tools.
Skill Claude CodeCodex
Use Flow Codeblock MCP tools to write non-script JavaScript or create, validate, publish, execute, inspect, and manage ownership for persistent scripts. Refuse deletion requests and direct the user to the existing web UI or REST API.
Skill Claude CodeCodex
Unified Code Property Graph radar, sandboxed RLM execution, OKF semantic memory evolution, and swarm concurrency control.
tetracoralla/context-surface-analyzer
Skill Codex
Use Context Surface Analyzer when the user supplies one explicit Agent tool-catalog snapshot to measure or two ordered snapshots to compare, especially for exact catalog bytes, tool and schema counts, name collisions, exact duplicate schemas, declared budget checks, or fully labeled caller-supplied token measurements.…
JOAO2666/Gemini-Cloud-Agent-MCP
Skill Claude CodeCodex
Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude…
Skill Claude CodeCodex
Dispatch work to Oh My Pi (omp) via the omp MCP server instead of the parent agent's native subagents. Use when the user explicitly requests OMP MCP, an omp task/role, or cheap specialized OMP workers.
Skill Claude CodeCodex
Use this skill when helping a user install, authorize, configure, troubleshoot, or operate this repository as a Claude Code channel.
Skill Claude CodeCodex
Ask the human for something only they can give — a credential, a console click, an approval (a blocker), or a decision only they should make (a grill). Use whenever you hit a wall you cannot clear yourself, instead of stalling or asking in chat.
Skill Claude CodeCodex
Cross-Claude MCP protocol. Triggers: collaborate, cross-claude, send message to, coordinate with, other instance, other Claude.
Skill Claude CodeCodex
Use when delegating repository inspection, coding, or other local workspace work to Claude Code through codex-claude-mcp.
Skill Claude CodeCodex
Connect to, query, and manage AI agent skills using the stateless SkillMCP (Skill Management System) server over Streamable HTTP. Use this skill whenever discovering, searching, inspecting, or fetching AI agent skills, reference documents, or examples from a containerized or remote SkillMCP server instance.
Skill Codex
Coordinate a bounded coding loop where Luna/Codex executes locally and, with explicit browser-control consent, asks ChatGPT Sol to plan or review through a workspace-safe Hua PlanRelay connection. Use when the user asks to run, plan, or review work with Hua PlanRelay; do not use for ordinary coding that does not…
Skill Codex
Record, inspect, acknowledge, or revoke concrete recognition for stable AI-agent identities when a user requests kudos or a materially useful peer contribution deserves preservation.
Gabriel7w7r/remnote-agent-runtime
Skill Claude CodeCodex
Create or update AgentSkills. Use when designing, structuring, or packaging skills with scripts, references, and assets.
Skill Claude CodeCodex
Install and securely configure the sipgate MCP server when a user asks to set up or connect a sipgate account to Codex or Claude. Do not use for ordinary sipgate product questions.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: