Find agent skills by outcome

大型语言模型驱动的个性化网络代理

发布时间：2024年10月22日

Tracing Smolagents

![Smolagents tracing via autolog](/docs/latest/assets/images/smolagents-tracing-ce4f2c411221b1c5d54a65e704e6a945.png)

import Admonition from "@theme/Admonition";

Agents are components that use reasoning to make decisions and take actions, designed to autonomously perform tasks or provide services with some degree of agency. LLM chains can only perform hardcoded sequences of actions, while agents use LLMs to reason through which actions to take, and in which

cli/services/agents.js - Agent Management Service

Provides agent lifecycle management including enabling, disabling, and resolving agent configurations. Handles agent aliasing, run modes (isolated/global/devel), and workspace structure creation.

MineNPC-Task: Task Suite for Memory-Aware Minecraft Agents

generic skill

Development Guidelines

- **Code Quality**: Write clean, readable, and maintainable code. Use meaningful variable and function names.

SWEPolyBench A multilanguage benchmark for repository level evaluation of coding

generic skill

AutoGPT: Automating Tasks with AI Agents

generic skill

Agents.md

This document provides instructions for AI agents working on this codebase.

Custom Agents Guide

Custom agents allow you to define specialized AI assistants with specific behaviors, tools, and prompts.

AGENTS.md: Drasi CLI

This document provides a high-level overview of the Drasi Command Line Interface (CLI) project for coding agents. The CLI is a Go application built with the Cobra library, responsible for installing and managing Drasi environments and resources.

h2oGPT integration with LangChain Agents

Various agents from LangChain are included:

tools Directory

This directory contains supporting files for the Sideload skill.

Tutorial: Using Agent Task Framework for Background Agent Instructions

The Agent Task Framework helps transform ambiguous user requests into clear, structured instructions that background agents can execute successfully. This tutorial shows practical examples from real project contexts.

Agents

generic skill

The is no unique solution but there are some approaches to build agents of increasing complexity. You could try to encode your knowledge, but that could be either too complex or you could be just not

generic skill

Virtual vs Non-Virtual Agents

There can be two types of Agents in Data Streams; Virtual and Non-Virtual Agents. When packaging an Agent using the Stream Integration Manager, it is important to specify if an Agent is Virtual by making sure that the _Virtual_ checkbox is correctly ticked.

SETUP_PARALLEL_AGENTS.md

**Claude Code: Please execute all steps below to set up parallel AI agents for the ZMF Production Dashboard**

Commands, Skills and Agents

generic skill

Quick Start: Activate Your Agents 🚀

Your agents are currently **IDLE** because they haven't been added to a flow yet. Here's how to get them working:

📋 Complete Guide: How to Assign GitHub Copilot Coding Agents to Issues

This guide covers all methods to assign GitHub Copilot Coding Agents to RFC implementation issues, from manual assignment to fully automated workflows.

AI Agents

generic skill

Resume Finished Agents Feature

Implemented the ability to resume finished agents **and agents that hit errors** by sending them a message. Previously, when an agent finished or hit an error (like rate limits), it would completely stop and could not be resumed. Now, these agents wait for user messages and can continue their work.

PitWall A.I. — AI Agent System Technical Documentation

> **Autonomous Multi-Agent System for Real-Time Race Analytics and Strategy Optimization**