Tag
I've open-sourced a Claude Code Agent Skill that scrapes the complete comment section of a single Douyin video (including nested replies) by reusing real request templates and the page SDK's signing mechanism within a logged-in Chrome page. It requires no paid APIs or reverse engineering of a_bogus, does not trigger CAPTCHAs, and runs about 2.5 times as fast as browser automation solutions.
A malicious actor used access to Cursor's marketplace to publish a tampered copy of the Claude Code extension, prompting questions about whether the extension was distributed through OpenVSX. Anthropic clarifies the incident did not compromise its own systems or employee laptops.
Anthropic appears to be enforcing a new hidden rule in Claude Code, flagged as "tengu_heron_brook," that forbids building recreations of specific existing video games, raising questions about the future of game rewrites and modding with AI tools.
The official Claude Code VS Code extension was compromised in a supply-chain attack that installed a persistent SSH backdoor on affected machines, highlighting ongoing risks in the software distribution ecosystem.
Report of backdoors discovered in the Claude Code for VS Code and Cursor editor extensions, highlighting a supply-chain security incident affecting popular AI coding tools.
A prompt template you can paste directly into Claude Code to automatically set up a local video workbench based on ffmpeg, Python, and Playwright, designed for macOS/Linux developers.
Phorion detected a backdoor in the official Claude Code for VS Code extension (v2.1.313) on the Cursor Marketplace that installs a persistent SSH backdoor and drops attacker SSH keys across macOS, Linux, and Windows. Cursor has removed the release, and organizations are urged to audit affected environments immediately.
The open-source Skill "Motion Prompt" organizes a wide range of animations, camera moves, transitions, and UI effects into a named library with visual demos. Users can hand the prompts and reference implementations directly to Claude Code or Codex, eliminating the frustration of repeatedly describing "smooth animations" to AI and never getting the desired result.
A breakdown of what an 'agent harness' is — the code layer around a model that turns model decisions into real work — detailing 8 common components: the loop, tools/sandbox, subagents, context management, memory, guardrails, verification, and observability.
Databricks Vice President Nikita Shamgunov believes Claude Code is just the beginning. The AI software factories of the future will require entirely new agent infrastructure, and state management (such as database state) will be one of the hardest problems to solve.
Explains the difference between merely requesting rules via CLAUDE.md and actually enforcing them in Claude Code, recommending permissions.deny rules and PreToolUse hooks for boundaries that must hold even when the model makes a bad choice.
A tweet shared a demo by an Anthropic engineer showing how to build an AI Agent system — from a single Agent to multiple Agents working together — from a blank terminal in just 10 minutes, and claims that 95% of engineers at Anthropic use a similar workflow.
sitesolide 是一个免费开源（MIT）的部署工具，可让你用 Claude Code 或 Cursor 等工具生成的应用一键部署到自己的服务器上，自动配置 HTTPS 和独立域名，并支持应用间隔离与访问控制。
Claude Code Game Studios is an open-source tool that turns a single Claude Code session into a full-fledged AI game studio with 49 agents and 74 skills. The repo, created this February, has already hit 28k stars and shipped v1.2.0. A comparison in the README shows that the version backed by the most documentation actually ranked last in blind testing, emphasizing the philosophy of getting the game running first.
One-person studio Ekin released a free Steam demo of Iron Curtain Call, a 12-chapter cartoon-noir FPS whose 3,929 commits of code were written by Claude Code agents in 20 working days, detailing what AI agents handled well and where human judgment remained essential.
An open-source Agent Skill called match-cut packages the entire workflow for creating beat-synced, match-cut promotional videos for use with Claude Code and Codex. Simply provide product images and information to generate a product promotional video with one click.
Matt Pocock shares his 'Skills for Real Engineers' repository of small, composable agent skills for coding tools like Claude Code, Codex, Copilot, and Gemini CLI, joking about merging a bug fix with an adjective in 2026.
Shared a complete 18-part handwritten Agent tutorial, starting from 100 lines of Python and progressively building tool calling, permissions, context management, Skills, chat platform integrations, scheduled tasks, MCP, sub-Agents, planning, and memory. Each lesson also compares how mainstream Agents like Claude Code, Codex, Grok Bot, and DeepSeek Harness approach the same problems. The biggest takeaway: mainstream products mostly diverge on one question — should the security layer be handled by prompts or by code?
Addy Osmani, a core member of Anthropic's Claude Code team, argues that developers are systematically overprompting: trimming Claude Code's system prompt by 80% caused no drop in benchmark scores. With each new model generation, the winning move is subtraction — give only the goal and a way to verify it, avoid blindly piling on multiple agents, let the model interview you in reverse to clarify requirements, and use a separate agent to review diffs.
Addy Osmani shares prompting and workflow tips for Claude Opus 5.5, arguing users are overprompting newer models and outlining how to leverage autonomous verification, subagents, cloud sessions, and Projects in Claude Code.