Category: tech > ai > agents

Comparison of three CLI-based browser agents for Claude Code: Playwright CLI (lowest token usage), Firecrawl Interact (browser automation plus scraping), and...

The .claude/ folder is the control center for Claude Code's behavior, containing instructions, permission rules, custom commands, skills, and subagent person...

Stripe ships roughly 1,300 agent-authored PRs per week, enabled not by the agent harness itself but by a decade of investment in cloud dev environments, inte...

PipesHub is an open-source, self-hostable workplace AI platform that unifies search across Google Workspace, Microsoft 365, Slack, Jira, Confluence and more ...

A developer reverse-engineered Claude Code's binary and discovered a fully built but server-gated feature called speculative execution - Claude predicts your...

Claude Code now supports scheduled autonomous tasks that run on a cron-like schedule in the cloud, with access to GitHub repos and MCP connectors. This enabl...

Cloudflare's Dynamic Worker Loader spins up V8 sandboxes in milliseconds, making per-request AI agent isolation 100x faster than containers.

GSD solves context window degradation by breaking work into atomic plans executed in fresh subagent contexts with parallel wave execution.

Claude Code's Dream feature automatically cleans memory files by resolving contradictions, merging duplicates, and removing stale information.

Everything Claude Code is an optimization system with 28 agents, 125 skills, and 60 commands that can cut AI coding costs by 60% or more.

CLI-Anything auto-generates CLI wrappers for desktop software like GIMP and Blender, making them controllable by AI agents like Claude Code.

NanoClaw runs Claude Code in isolated containers connected to WhatsApp and Telegram, with only 15 source files and 3,900 lines of code.

ProofShot records video and screenshots while AI agents interact with browsers, packaging everything into standalone HTML proof artifacts.

Claude Code's YOLO mode paired with GSD and Superpowers skills enables running multiple autonomous sessions in parallel without manual approval.

Three Claude Code skills for productivity: GSD for execution, Superpowers for capabilities, and Create Skill for capturing reusable workflows.

A Claude Code skill that generates complete design systems from project descriptions, covering 161 product categories and 67 UI styles.

Anthropic shipped Remote, Schedule Tasks, Dispatch, and Channels for Claude Code, replicating the autonomous-agent workflow OpenAI promised.

A Claude Code skill bridges to Notebook LM for generating audio, video, and slides, then adapts the outputs to match your brand voice.

Paperclip is a multi-agent framework where a CEO agent auto-creates and coordinates a team of AI agents that self-onboard to your codebase.

Spec-driven development tools create excessive review overhead and a false sense of control, echoing model-driven development's historical failures.

Agent Browser's Rust rewrite cuts install size from 710 MB to 7 MB, uses 18x less memory, and reduces token usage by 93% via snapshot commands.

Enabling LSP in Claude Code gives 100x faster code search than grep, returning exact definitions in 50ms while burning far fewer tokens.

Claude Dispatch lets you control Claude Co-Work from your phone, sending tasks remotely while Claude works with your local files and apps.

Cook adds review loops, parallel racing, and task-list progression on top of AI coding agents like Claude Code, Codex, and OpenCode.

Stripe's Machine Payments Protocol lets AI agents pay via HTTP 402 responses, supporting both fiat and stablecoins. Submitted to IETF as a standard.

LinkedIn's editor shipped iOS apps to the App Store with zero coding experience, using a dual-agent Claude Code workflow for building and review.

Simon Willison breaks down how coding agents work: an LLM in a loop with tools, token caching, and reasoning modes. Simpler than you think.

Most Claude Code Skills fail at triggering, not execution. Writing a specific description with explicit trigger keywords matters most.

OpenViking organizes AI agent context as a virtual filesystem with tiered loading, cutting tokens 11x while raising task completion by 15 points.

Simon Willison defines agentic engineering as building software with coding agents in a loop, distinct from unreviewed vibe coding.

CoPaw is a self-hostable AI assistant from Alibaba that connects to messaging platforms and supports local LLMs via Ollama with no API key needed.

Context Mode achieves 94-100% context reduction per tool call, extending typical Claude Code sessions from 30 minutes to about 3 hours.

A multi-agent LLM workflow uses Opus for architecture, Sonnet for coding, and Codex/Gemini for review. Different models catch different failures.

NanoClaw replaces OpenClaw's 500k lines of code with a handful of files and hypervisor-level container isolation per AI agent.

A concise ruleset for AI agent workflow orchestration covering plan mode, subtask strategy, self-improvement loops, and verification discipline.

The Playwright CLI is roughly 500% more token-efficient than the Playwright MCP for browser automation in Claude Code, with the same capabilities.

A promotional TikTok claims CLI tools will replace MCP servers for Claude Code, but provides no evidence. The actual repo may have more substance.

Google's Nano Banana Pro 2 integrates web search into image generation, producing grounded outputs with accurate text from live data.

VS Code Agent Kanban stores every AI coding task as a markdown file with YAML frontmatter, solving context loss between sessions.

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