#agents

29 items tagged #agents.

GUIDE 2026-05-22

Running scheduled coding agents safely

Antigravity 2.0 and Anthropic's Routines let agents run unattended on a schedule or a webhook. With no human at the trigger, scope, a review gate, a cost cap, and a kill switch are the only controls that still work.

Owner · 4 min #security #agents

BLOG 2026-05-22

The agent that runs while you sleep

Antigravity 2.0 schedules agents to run in the background, and Anthropic's Routines have done it since April. The supervision model we built for AI coding assumed someone was watching each run. Scheduled agents quietly remove the watcher.

Owner · 6 min #opinion #agents

GUIDE 2026-05-20

Running Cursor parallel agents safely

Cursor 3.5's multi-repo automations let one developer run agents across several codebases at once. Each running agent is a separate diff you have to review. Here is how to bound the concurrency, cap the spend, and avoid rubber-stamping.

Owner · 4 min #cursor #agents

BLOG 2026-05-20

Parallel agents and the per-developer meter

Cursor 3.5 lets one developer run agents across several repositories at once, and Composer 2.5 makes each call cheaper. Both move the same number in the same direction: how many tokens one person burns in a day.

Owner · 6 min #opinion #agents

NEWS 2026-05-20

Cursor ships Composer 2.5 and multi-repo agents

Cursor released its Composer 2.5 model on May 18 and Cursor 3.5 on May 20. The model gets cheaper Standard token rates; the editor gets automations that run agents across several repositories at once.

Owner · 2 min #cursor #release

GUIDE 2026-05-19

Hardening your agent's supply chain

A concrete checklist for the May 2026 failure modes: pin skills and MCP servers, SBOM the agent's dependencies, require auth on MCP, and review config files as code.

Owner · 4 min #security #supply-chain

GUIDE 2026-05-19

Measuring your own AI-code share honestly

Why 'percent of lines AI wrote' is a vanity metric, what to measure instead, and a lightweight git-based method that counts work which survived contact with review.

Owner · 4 min #productivity #measurement

BLOG 2026-05-19

Supervision from your pocket

OpenAI just made approving an agent's work something you can do from a phone in a checkout line. That is exactly the thing the supervision paradox warned about, now with great UX.

Owner · 6 min #opinion #agents

BLOG 2026-05-19

The 60% claim, deconstructed

Airbnb's CEO said AI writes 60% of the company's new code. The number is real and the conclusion people draw from it is not. This is the gap between the two.

Owner · 6 min #opinion #productivity

BLOG 2026-05-19

Your agent's config is the attack surface now

I asked the MCP supply-chain question a while ago without numbers. May 2026 supplied the numbers — and they say the breach is mundane, not exotic, and the fix is boring.

Owner · 6 min #opinion #security

NEWS 2026-05-19

A wave of agent-targeting CVEs lands in May 2026

Researchers disclosed a cluster of vulnerabilities in AI coding agents in May 2026: prompt-injection-to-RCE in CrewAI, an unauthenticated Azure SRE Agent endpoint, a poisoned AGENTS.md chain against Codex, and over a thousand malicious skills.

Owner · 2 min #security #agents

GUIDE 2026-05-18

Getting started with Grok Build's plan mode

A first session with xAI's Grok Build: install, generate and edit an execution plan before any code is written, review the diff, and run it safely alongside the agent you already use.

Owner · 4 min #grok-build #agents

GUIDE 2026-05-18

Using AI coding tools without letting your skills atrophy

Practical habits to keep debugging, architectural judgment, and code-reading sharp while still using agentic tools every day: selective delegation, read-before-accept, and scheduled unaided work.

Owner · 4 min #skills #agents

BLOG 2026-05-18

The coding-agent race gets crowded

xAI's Grok Build is the fourth major terminal coding agent. Its shape is nearly identical to the others — and that sameness is the most interesting thing about it.

Owner · 6 min #opinion #agents

BLOG 2026-05-18

The supervision paradox

Lars Faye's 'Agentic Coding is a Trap' and Anthropic's own 47% finding point at the same loop: supervising an agent needs the skills agent use erodes. I think the loop is real and the fix is boring.

Owner · 6 min #opinion #agents

NEWS 2026-05-18

xAI ships Grok Build, its first coding agent and CLI

xAI opened the Grok Build beta on May 14 for SuperGrok Heavy subscribers. The terminal agent runs up to eight workers in parallel, carries a 2M-token context window, and ships with an approve-first plan mode.

Owner · 2 min #grok-build #release

GUIDE 2026-05-14

Running parallel coding sessions with Claude Code Agent View

Claude Code's Agent View lets you run multiple sessions simultaneously without terminal tab-juggling. Here is how to start background sessions, check their status, and decide when parallel sessions actually help.

Owner · 4 min #claude-code #agents

NEWS 2026-05-14

Claude Code ships Agent View: one dashboard for every session

Anthropic's May 11 update adds a unified session list to Claude Code, letting you track, background, and switch between multiple concurrent coding sessions from the terminal.

Owner · 2 min #claude-code #release

GUIDE 2026-05-08

A working Windsurf Cascade workflow for multi-file features

Cascade is Windsurf's agent. It can write 200 lines across 6 files in one go, or it can produce a tangled mess. The difference is how you brief it.

Owner · 9 min #windsurf #cascade

NEWS 2026-05-06

Cursor Background Agents go GA: run tasks while you keep coding

Cursor's background agent feature exits beta. Agents can now run terminal commands, edit files, and report back without blocking your main editor session.

Owner · 3 min #cursor #agents

NEWS 2026-05-05

GitHub Copilot Workspace exits preview: pricing, capabilities, and the comparison most people will run

Copilot Workspace, GitHub's task-oriented AI environment, is now generally available. Here's what's in the box, what isn't, and how it stacks up against Cursor's agent features.

Owner · 5 min #copilot #workspace

NEWS 2026-04-16

Cline ships first-class MCP server support: agents that can talk to your tools

Cline now supports the Model Context Protocol natively. Configure an MCP server, and Cline can call it as a tool — connecting the agent to your databases, APIs, and internal services.

Owner · 5 min #cline #mcp

BLOG 2026-03-20

AI agents still can't do good frontend, and the reasons are interesting

Agent loops handle backend code reasonably well. They produce serviceable but bland frontends, and the gap reveals something about what 'frontend work' actually is.

Owner · 5 min #frontend #agents

NEWS 2026-03-10

OpenAI's Codex CLI enters public beta with terminal-first agent workflow

OpenAI launched a CLI agent product. The shape is similar to Claude Code: terminal-based, agentic, with file-system and shell access.

Owner · 4 min #openai #codex

NEWS 2026-02-15

Cursor's Shadow Workspace lets agents experiment without touching your files

Cursor 1.1 adds Shadow Workspace, a feature that runs agent operations in a hidden copy of your project. Implications for safer agent workflows.

Owner · 3 min #cursor #agents

BLOG 2026-02-10

Agentic coding and the end of 'pure' code review

When AI agents produce most of a PR's code, what is review actually for? The answer is shifting in important ways.

Owner · 4 min #code-review #agents

NEWS 2026-02-02

Google opens 'Jules' coding agent to public beta

Google's coding agent enters public beta with GitHub integration and Gemini 2.0 backing. The major-lab agent space gets another credible entry.

Owner · 3 min #google #jules

BLOG 2026-01-25

Why agent mode doesn't help me debug, even when it should

Agent loops are designed for autonomous problem-solving. For debugging, the autonomy is precisely what hurts. Investigation requires staying in the loop.

Editor · 4 min #debugging #agents

BLOG 2026-01-08

The rise of personal coding agents: tools you build for yourself

Custom coding agents built for specific personal workflows are becoming a real category. The pattern, the tooling, and what's working.

Owner · 4 min #agents #workflow