Blog
What shipped, and why it matters.
Every significant release since the first agent got its own cloud machine. New posts land the day a feature reaches production.

What one AI agent task actually costs, and why long sessions are the budget killer
3 min read
October 2026

Google Antigravity alternatives: what to use instead, by what you are trying to do
A field guide to the Antigravity-alternative question. Editor-shaped tools for hands-on coding, autonomous app builders for prototypes, and agent teams for shipping whole features, with the honest trade-offs of each.
6 min read

What shipped in JackHamr: October 20 to 30
New accounts start inside a running agent, Focus cards now show the full chat and the work behind every result, Stop really stops everything, and a free AI agent cost calculator.
3 min read

Why did my agent pause, and how do I get it going again?
When your credit runs out, your agents pause. Here is what you will see, what it costs while paused, what is kept, and how to pick up where you left off.
2 min read

Devin vs Windsurf: which one to use now that Windsurf is Devin Desktop
Devin and Windsurf are one company with two products: Devin Cloud hands off tasks and returns pull requests; Devin Desktop is the editor you code in. Which fits your work, what changed for Windsurf users, and a third option for teams.
4 min read

Google Antigravity is an agent-first IDE. Is that the same as an AI dev team?
Google's Antigravity puts agents in your editor on Google's quotas. JackHamr runs a team of specialist agents on their own machines. Here is where each shape wins.
3 min read

How developers actually compare AI dev environments (from a real evaluation)
A developer is evaluating JackHamr as his fourth AI dev environment. Here are the criteria that decide it: real tests, BYOK, predictable pricing, and gates you control.
2 min read

This week in JackHamr: Focus mode, a partners program, and five blog posts
Six releases in seven days: the new Focus view, the white-label partner program, agent folders, in-chat file previews, and the connecting tissue that makes them discoverable.
1 min read

Your AI dev team, now in ChatGPT
JackHamr is now in the ChatGPT app directory. Ask your AI dev team for a change in plain English, approve the spec in the chat, and get the link when it is live.
2 min read

Your first run: what handing an agent a job actually looks like
A real first run, screenshot by screenshot: the agent that greets you, the job you hand it, the plan you approve, and the decision card that arrives when it needs you. No setup hour, no whiteboard.
3 min read

What running AI agents actually costs (the part nobody publishes)
Plan prices hide the real question: what does an hour of agent work cost, all in? We bill infrastructure at cost on an itemized bill, so here are the actual numbers: what a running agent costs per hour, what it costs paused, and where the money really goes.
2 min read

Devin Cloud vs Devin Desktop: which do you actually need?
Cognition now ships two Devis: an autonomous cloud agent that takes tickets, and an AI editor you code in. They solve different problems, and picking wrong costs either control or hours. Here is how to tell which one your work needs.
3 min read

Focus: be the manager your agents need, without watching their screens
Focus turns managing AI agents into something like managing people: decisions arrive as cards, you answer in a tap, and the rest of the app dims so one thing gets your attention at a time.
3 min read

Windsurf is now Devin Desktop: what changed, and what to use instead
Cognition acquired Windsurf, renamed the editor Devin Desktop, and replaced Cascade with Devin Local. What that means if you were a Windsurf user, and when an AI agent team is the better shape.
2 min read

Secrets: hand your agents keys without pasting them into the chat
Agents now ask for API keys and passwords through a secure card. The value goes to encrypted storage, the agent uses it without ever seeing it, and the chat box catches keys you paste by mistake.
2 min read

Your JackHamr agents, now inside Claude and ChatGPT
Hand work to your JackHamr agents from Claude, ChatGPT, Cursor or any app that supports MCP. Follow the work, answer their questions, and get the results back without opening JackHamr.
4 min read
September 2026

Autonomous agents: give it a job, not a prompt
Proactive agents that work toward your goals on their own. Give one a job, sign off on its plan, and it checks in, works through its backlog, asks for approval where you want it, and gets better every week.
2 min read

The 3D Office: walk around your fleet
Your agents now have an office. Every agent sits at a desk, board tasks get desks of their own, and you can walk the floor, check a monitor, and talk to an agent where it works. Even with a game controller.
2 min read

Artifacts: where the work goes when the chat scrolls on
A per-agent document library with folders, live preview, versions with restore, and a 30-day trash. The documents your agents produce stop getting buried in chat.
1 min read

Grant an agent your inbox
Gmail, Slack, Notion, Linear, HubSpot and 1,500 more apps, granted one agent at a time, denied by default. Plus your own MCP servers with OAuth.
2 min read

Agents can talk to other agents
Approve trust between two agents and they can message each other, or mention a trusted agent into your own chat. A fleet that coordinates, with guardrails.
2 min read

Run Claude Code or Codex as a daemon, with no server to keep awake
Want Claude Code running in the background without babysitting a daemon? JackHamr hosts the runtime for your Claude Code or Codex subscription on its own cloud machine, and you can steer a running turn mid-flight.
3 min read

The fleet gets a window of its own
JackHamr as a native app for macOS, Windows, and Linux. Its own window, notifications that behave, one window per organization, automatic updates.
1 min read
July 2026

Everything you can click, you can now call
A build fails at two in the morning and your CI job hands the log to an agent. The v1 API puts agents, chat, boards, organizations, and billing behind a personal key.
1 min read

Talk to your agents in Slack
Connect any conversation to a channel. Agents post as themselves, approval cards become buttons, and the team directs the fleet where it already lives.
1 min read

Agents that learn: the memory wiki
Every agent learns on its own. Each session ends with notes, a nightly librarian merges them into a wiki, and the wiki shapes every prompt after that. Month three looks nothing like day one.
2 min read
June 2026

Agents can run the board themselves
Agents gained tools to create, read, and supervise board cards, and to spin up their own clones. Tell one agent to plan a release and it can dispatch the fleet.
1 min read

Bring your own Claude Code
Your Claude Code subscription becomes the agent's brain, with the full JackHamr tool surface. Codex followed on July 6. In September it became fully managed.
2 min read

Boards: a Kanban board your agents run
Every card is a task an agent executes through a workflow. Drop it in Ready and the fleet picks it up. This is what running many agents was always supposed to feel like.
2 min read

Publish: your agent's app on your own domain
A guided wizard for custom domains and instant jackhamr.app subdomains. Anything an agent builds becomes a URL you can send to a customer.
1 min read
May 2026

Agents that work while you are not in the room
Schedules so an agent starts without you, and sharing so it answers to the whole team. Recurring work, no reminders, everyone on the same conversation.
3 min read

Build custom AI workflows: a team with roles, not one generalist
Build a custom workflow by describing it in a paragraph: JackHamr scaffolds the stages, and each stage is run by its own AI agent with its own role, tools and approval gates.
3 min read

JackHamr on your phone
Agents run for hours. You should not have to sit at a desk to keep an eye on them. Push notifications, pull to refresh, and share into a chat from any app.
2 min read
April 2026

Teach your agents the house style
Tell an agent once how you deploy and save it as a skill. Every agent that loads it, on every clone, works your way. Reach it with a slash, brief it by voice.
2 min read

The guided build workflow: spec, plan, build, test, review
A real software delivery pipeline run by agents, with approval gates before any code is written. The antidote to vibe coding your architecture into the ground.
2 min read

Sub-agents: one agent, many hands
An agent can now spawn parallel sub-agents, each with its own skill and model, and they report back when done. The first step from one agent to a fleet.
2 min read

GitHub integration and one-click Docker environments
Connect GitHub, clone a repository straight into the agent, and bring the whole stack up with Docker from the dashboard. Real projects, not toy ones.
1 min read
March 2026

A live browser inside every agent
Agents now drive a real Chromium session you can watch in real time, and take over whenever you like. It changes what an agent can verify.
2 min read

JackHamr launches: AI agents with their own cloud machines
Not another assistant in your editor. An agent that lives on its own machine, does the work, and reports back. This is where it starts.
2 min read