skills.rest
tendercoconut174
tendercoconut174COMMUNITY
Explore all skills by this author →
GitHub

agent-skill

Generate CrewAI agents, system prompts, and agent registry entries for LangGraph workflows.

Updated Mar 12, 2026
Search Tags:#llm#multi-agent#langgraph#system prompts#agent generation#crewai#tool registry
One-click install
npx skills add https://github.com/tendercoconut174/ai-agent-platform --skill agent-skill-tendercoconut174
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-skill
Source: https://github.com/tendercoconut174/ai-agent-platform/tree/main/.cursor/skills/agent-skill
Command: npx skills add https://github.com/tendercoconut174/ai-agent-platform --skill agent-skill-tendercoconut174

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Building and registering new AI agents in a multi-agent platform involves coordinating CrewAI crew definitions, tool bindings, and registry entries across several files, which is error-prone when done manually. ## Core Features & Use Cases - Agent Generation: Create new CrewAI sub-agents (research, analysis, generator, code, monitor) with role, goal, and backstory definitions in _CREW_DEFS. - Registry & Tool Binding: Register agents in AGENT_REGISTRY and assign LangChain tools via TOOL_REGISTRY with automatic code-approval detection. - Use Case: When you need a new monitoring agent that watches a data source, use this Skill to generate the crew definition, runner function, registry entry, and tool assignments following the platform's established patterns. ## Quick Start Use the agent skill to create a new CrewAI research agent with web_search and scrape_url tools and register it in the agent registry.

Frequently Asked Questions about agent-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a new CrewAI agent in a LangGraph workflow?▼

Add an entry to _CREW_DEFS in crewai_agents.py with role, goal, and backstory, then create a run_<type> runner function. Register it in AGENT_REGISTRY and assign tools through TOOL_REGISTRY in the MCP server.

How do I bind tools to a CrewAI agent?▼

Tools are LangChain @tool wrappers assigned per agent type in TOOL_REGISTRY within shared/mcp/server.py. They are converted to CrewAI tools via Tool.from_langchain() when the crew is constructed.

What is the difference between CrewAI sub-agents and special agents?▼

CrewAI sub-agents (research, analysis, generator, code, monitor) run as single-agent crews inside LangGraph. Special agents like chat, plan_execute, and scheduler use their own implementations without CrewAI.

Can agents communicate directly with users in a multi-agent system?▼

No, agents never talk to users directly. All results return to the Supervisor through the execute node, which evaluates outcomes and routes final output through the Delivery Service.

How is code approval handled for agents that execute Python?▼

Agents with tool_execute_python in their tool set are automatically treated as code-approval agents when require_code_approval is true. This is derived from TOOL_REGISTRY rather than hardcoded agent lists.

Related Skills

Explore More Software Engineering Skills→
knowz-ioknowz-io

kc-register

Register a KnowzCode user and configure the knowz MCP server for Gemini CLI.

Official
Intermediate
taewook486taewook486

agency

Route user requests through specialized agents for website and app production workflows.

Community
Advanced
iamtouchskyeriamtouchskyer
189

opc

Coordinate a multi-agent digraph review pipeline with gate verdicts.

Community
Advanced
marvelousempiremarvelousempire

agent-orchestration-frameworks

Guide framework selection via decision trees for agent orchestration projects.

Community
Advanced