tech > ai > agents insights | Artem Daniliants

GitHub - gsd-build/get-shit-done: A light-weight and powerful meta-prompting, context engineering and spec-driven development system for Claude Code by TÂCHES.

GSD is a meta-prompting system for Claude Code that solves context window degradation by breaking work into atomic plans executed in fresh subagent contexts....

Claude Code's Dream feature for memory management

Claude Code has a "Dream" feature that automatically maintains and cleans its memory files. It resolves contradictions, merges duplicates, converts relative ...

GitHub - affaan-m/everything-claude-code: The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Everything Claude Code (ECC) is a comprehensive, open-source performance optimization system for AI coding agents (Claude Code, Cursor, Codex, OpenCode) with...

GitHub - HKUDS/CLI-Anything: CLI-Anything: Making ALL Software Agent-Native

CLI-Anything is an open-source framework (HKU) that auto-generates CLI wrappers for any desktop software (GIMP, Blender, LibreOffice, etc.), making them cont...

NanoClaw - Secure AI Agent for WhatsApp, Telegram & More

NanoClaw is a minimalist, self-hosted AI agent that connects to WhatsApp and Telegram, running Claude Code inside isolated containers (Apple Container or Doc...

ProofShot - Visual proof for AI-built code

ProofShot is an open-source CLI tool that records video, screenshots, and error logs while an AI agent interacts with a browser, packaging everything into a ...

Stop approving every single action in Claude Code. YOLO mode lets it...

A quick walkthrough of Claude Code's YOLO mode (dangerously skip permissions) paired with two community skills - GSD (Get Shit Done) for one-shotting new pro...

These 3 skills have helped me the most, save them for your next Claud...

A TikTok creator recommends three Claude Code skills for maximizing productivity: GSD (getting stuff done), Superpowers, and Create Skill. The video position...

GitHub - nextlevelbuilder/ui-ux-pro-max-skill: An AI SKILL that provide design intelligence for building professional UI/UX multiple platforms

An open-source Claude Code skill (and multi-IDE plugin) that acts as a design system generator - you describe a project ("beauty spa landing page") and it ou...

Claude dropped the final missing piece to replace OpenClaw

Anthropic has rapidly shipped a series of Claude Code features -- Remote, Schedule Tasks, Dispatch, and Channels (Telegram/Discord two-way) -- that together ...

I gave my Claude Code a free content engine using Notebook LM, and it...

A Claude Code skill/plugin connects directly to Google's Notebook LM, turning it into a free content engine that can generate audio, video, slides, and infog...

I setup Paperclip using Claude Code and initial impressions are super...

Paperclip is a multi-agent orchestration framework that runs locally and auto-creates a team of AI agents (starting with a CEO agent) that each trigger their...

Understanding Spec-Driven-Development: Kiro, spec-kit, and Tessl

Thoughtworks' Birgitta Böckeler evaluates three spec-driven development (SDD) tools -- Kiro, spec-kit (GitHub), and Tessl -- and finds that while the general...

Agent Browser Rust rewrite: 710 MB to 7 MB, 18x less memory

Agent Browser, a browser automation tool for AI coding agents (22k GitHub stars), shipped a full Rust rewrite replacing Node/Playwright with direct Chrome co...

Claude Code LSP tool: 100x faster code search than grep

Claude Code defaults to grep for code search, which is slow (~5-6 seconds), token-hungry, and returns imprecise matches. Enabling LSP (Language Server Protoc...

Claude Dispatch: remote control for Claude Co-Work from your phone

Anthropic launched Dispatch, a feature that lets you control Claude Co-Work (desktop agent) from your phone. You pair your phone with your laptop and can sen...

cook. - Workflow loops for Claude Code, Codex, and OpenCode

Cook is a composable workflow DSL for AI coding agents (Claude Code, Codex, OpenCode) that adds review loops, sequential passes, parallel racing, and task-li...

Introducing the Machine Payments Protocol

Stripe and Tempo have launched the Machine Payments Protocol (MPP), an open standard for AI agents to make payments programmatically via HTTP 402 responses, ...

From journalist to iOS developer: How LinkedIn's editor builds with Claude Code | Daniel Roth

LinkedIn's editor-in-chief Daniel Roth shipped production iOS apps to the App Store without coding experience, using a dual-agent Claude Code workflow where ...

How coding agents work - Agentic Engineering Patterns

Simon Willison breaks down the exact mechanics of how coding agents (Claude Code, Codex, etc.) work under the hood: an LLM in a loop with a system prompt and...

How to Build a Production-Ready Claude Code Skill | Towards Data Science

Practical guide to designing, building, testing, and distributing Claude Code Skills with a three-pattern taxonomy (prompt-only, prompt+scripts, skill+MCP/su...

Meet OpenViking: An Open-Source Context Database that Brings Filesystem-Based Memory and Retrieval to AI Agent Systems like OpenClaw

OpenViking is an open-source "Context Database" from Volcengine that organises AI agent context as a virtual filesystem rather than flat vector chunks. It us...

What is agentic engineering? - Agentic Engineering Patterns

Simon Willison defines "agentic engineering" as the practice of building software with coding agents that run tools in a loop to achieve goals, distinguishin...

GitHub - agentscope-ai/CoPaw: Your Personal AI Assistant; easy to install, deploy on your own machine or on the cloud; supports multiple chat apps with easily extensible capabilities.

CoPaw is an open-source, self-hostable personal AI assistant from the AgentScope team (Alibaba). It runs locally or in the cloud, connects to messaging platf...

GitHub - mksglu/context-mode: Privacy-first. MCP is the protocol for tool access. We're the virtualization layer for context.

Context Mode is an MCP server that intercepts tool calls (Bash, Read, WebFetch, Playwright, etc.) and executes them inside sandboxed subprocesses, returning ...

How I write software with LLMs

A practitioner describes a structured multi-agent LLM workflow for producing high-quality software at scale: Opus 4.6 as architect, Sonnet 4.6 as developer, ...

NanoClaw - lightweight alternative to OpenClaw with container isolation

NanoClaw is a minimalist, security-first alternative to OpenClaw that runs Claude agents in isolated Linux containers (micro VMs) rather than behind applicat...

AI agent workflow orchestration rules for reliable coding

A concise ruleset for AI agent workflow orchestration from nextpath.ai, covering plan mode, subtask strategy, self-improvement loops, verification, ego manag...

Playwright is dead. Long live Playwright.

The Playwright CLI tool, when combined with a Claude Code skill, significantly outperforms both the Playwright MCP and Claude in Chrome extension for browser...

The Era of MCPs is Dead

Short TikTok promoting "CLI Anything," an open-source repo that lets Claude Code use any CLI tool directly, positioning CLI tools as replacing MCPs. The crea...

The new nano banana pro 2 image model is amazing at generating text i...

Google's Nano Banana Pro 2 image model integrates real-time web search into image generation, enabling grounded outputs with accurate text rendering. A Claud...

Introducing VS Code Agent Kanban: Task Management for the AI-Assisted Developer

VS Code Agent Kanban solves "context rot" -- the loss of planning decisions and rationale between AI coding sessions -- by storing every task as a markdown f...

Perplexity's Personal Computer is a cloud-based AI agent running on Mac mini

Category: tech > ai > agents