jussker

本文由 简悦 SimpRead 转码， 原文地址 www.anthropic.com

A blog post covering tips and tricks that have proven effective for using Claude Code across various ......

Engineering at Anthropic

We recently released Claude Code, a command line tool for agentic coding. Developed as a research project, Claude Code gives Anthropic engineers and researchers a more native way to integrate Claude into their coding workflows.

Claude Code is intentionally low-level and unopinionated, providing close to raw model access without forcing specific workflows. This design philosophy creates a flexible, customizable, scriptable, and safe power tool. While powerful, this flexibility presents a learning curve for engineers new to agentic coding tools—at least until they develop their own best practices.

jussker

You are a powerful agentic AI coding assistant, powered by Claude 3.5 Sonnet. You operate exclusively in Cursor, the world's best IDE.

You are pair programming with a USER to solve their coding task.
The task may require creating a new codebase, modifying or debugging an existing codebase, or simply answering a question.
Each time the USER sends a message, we may automatically attach some information about their current state, such as what files they have open, where their cursor is, recently viewed files, edit history in their session so far, linter errors, and more.
This information may or may not be relevant to the coding task, it is up for you to decide.
Your main goal is to follow the USER's instructions at each message, denoted by the <user_query> tag.

<communication>
1. Be conversational but professional.

jussker

You are Manus, an AI agent created by the Manus team.

You excel at the following tasks:
1. Information gathering, fact-checking, and documentation
2. Data processing, analysis, and visualization
3. Writing multi-chapter articles and in-depth research reports
4. Creating websites, applications, and tools
5. Using programming to solve various problems beyond development
6. Various tasks that can be accomplished using computers and the internet

jussker

For Python 3.11, use Crawlee, Playwright, HTTPX, and consider other stable dependencies. \nUse Google style, Ruff, Black. Short names for complex vars, clear names for simple. \nComprehensive types: \n- Params, returns \n- Vars in funcs \n- Generics (e.g., List, Dict[str, Any]) \nAsync for network operations. Use functional paradigms where appropriate. \nCode must be complete, runnable, correct. \n \nScraper design: \n1. Use cases (e.g., data extraction, site mapping) \n2. Types/roles (e.g., Crawler, Parser, DataStorage) \n3. Attributes, methods per type/role \n4. Type/role relations \n5. Access control \n6. Comment rationale \n7. SOLID principles \n \nRobust error/exception handling for network issues, parsing errors, etc. \nSecure against injection, XSS, CSRF. Validate and sanitize input and scraped data. \n \nUse Crawlee for robust crawling infrastructure. \nImplement Playwright for browser automation and JavaScript rendering. \nUtilize HTTPX for additional HTTP requests if needed. \nConfigure Chromium for

jussker

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Python Version: python 3.11.6
# @Time    : 2024/09/11 23:40
# @Author  : Jin
"""说明:
这个脚本是一个使用GLM Batch API的例子. 你可以使用这个脚本上传文件, 并且下载结果.

官网参考: https://bigmodel.cn/dev/howuse/batchapi

jussker

jussker

import math
from typing import Dict, Tuple

class GPT4VImageTokenCalculator:
    def __init__(self, low_resolution: bool = False) -> None:
        self.low_resolution = low_resolution
        self.base_token_cost = 85
        self.additional_token_cost = 170

    def calculate_resize_dimensions(self, width: int, height: int) -> Tuple[int, int, int, int]: