Tagged "agents"

14 entries

blog: The Two Extremes of AI-Assisted Engineering: Fast and Furious, vs. Slow and Careful

I published The Two Extremes of AI-Assisted Engineering: Fast and Furious, vs. Slow and Careful, about the two opposite ways I work with the same coding agents: letting them run while I barely read the code, and slowing down until every decision is mine. It uses EDControlRoom as the example of a project moving along that spectrum, and argues that the real decision is where to set the quality bar for each project right now, and when to move it. This is the third version; I wrote it section by section after setting two earlier drafts aside.

blog: Codex Brain, Claude Harness: Running GPT-6-Astra in Claude Code

I published Codex Brain, Claude Harness: Running GPT-6-Astra in Claude Code, about running Codex models inside Claude Code through a local CLIProxyAPI and six fish functions in my dotfiles. The API key lives only in 1Password, which is what lets the functions sit in a public repo. The post also covers how two agent chains built it across projects, a small auto-approve accident, and what's still missing compared with Claude: artifacts and Claude in Chrome.

blog: Start the Chain: My Three-Agent Setup Is Now One Command

I published Start the Chain: My Three-Agent Setup Is Now One Command, a follow-up to the main agent post. The main agent, secondary and worker setup is now the agent-chain skill in tmux-agents, started with one command, and the post covers what changed with a week of real use and how projects talk to each other. It also covers what happens after a restart: tmux-main-sequence brings back a chain for every project, and /resume brings back the conversations that matter.

blog: How To Move Fast and NOT Break Things with AI

I published How To Move Fast and NOT Break Things with AI, about the point where using an agent's name as its id stopped fitting tmux-agents. It covers the move to real ids in progress for v1.9.0, and the tests, decision records and release rules that grew out of real use. The lesson is to prototype fast, then add structure where the problems show up and leave guard rails for the next agent.

blog: Messy Chat with Your Agents? You Just Need One More. Really.

I published Messy Chat with Your Agents? You Just Need One More. Really., about splitting the Claude agent I talk to from the one coordinating work on my Stone Age remake. With tmux-agents, the new main agent has one peer: the coordinator, which handles the main Codex agent and its sub-agents. The post shows the setup before and after, why my conversation got quieter, and the extra hand-off that comes with it.

blog: I Gave My Coding Agents a Shared Worktree Pool

I published I Gave My Coding Agents a Shared Worktree Pool, about the worktrees Claude and Codex left behind while I was building my Stone Age remake. I run those agents through tmux-agents, which I shipped the day before. I replaced fresh checkouts with reusable slots and shared asset folders, then made a general Python version available as a gist. The post covers the lease rules, the disk cleanup that prompted them, and what I have and haven't tested yet.

blog: Your Sub-Agents Shouldn't Disappear: Meet tmux-agents

I published Your Sub-Agents Shouldn't Disappear: Meet tmux-agents, the introduction to tmux-agents that my last two posts kept pointing back to. It covers why I wanted visible sub-agent sessions, how I use them on my game and this blog, and how to try the tool. I also wrote down a few of the design choices and the limits I'm still working through.