6 posts tagged with "ai"

A Trick is the most boring primitive we have. It runs your container once, watches it finish, records the exit codes, and tears itself down. No routing, no Genie, no second act. So when someone proposed building "self-healing CI" on top of it, the obvious reaction was: with what? There's nobody home in a Trick.

That turned out to be the point. The Trick stays dumb; the intelligence lives elsewhere — a separate coding Genie that wakes up after the job is done, reads what went wrong, and tries to fix it. Compose a one-shot job with that out-of-band agent and you get a loop that runs your tests and, when they're red, takes a swing at making them green. We call the patterns CI-Job and Muti-Job, and the cleverness is never inside the Trick.

Your coding agent is great at writing the diff. It is much worse at the part that comes after: spin up an environment, wait for it to actually come up, hand you a URL, tail the logs when something breaks. Today it does that by shelling out to the fibe CLI and string-matching the help text — which works until the agent invents a flag that doesn't exist, or misreads "building" as "ready," and you spend ten minutes untangling what it thought it did.

The Fibe MCP server fixes the seam. The same fibe binary you already have doubles as a Model Context Protocol server, exposing the entire platform as typed tools an agent calls directly — no subprocess, no shell parsing, no guessing. This guide walks you through wiring it into Claude Code, Cursor, or Codex, choosing how many tools to put in front of the agent, and what it can actually do once it's connected. By the end you'll have an agent that can take "launch the rails template and give me the URL when it's up" and just do it.

The Fibe SDK is one Go binary doing two jobs. Type fibe playgrounds list and it's a normal Cobra CLI. Run fibe mcp serve and the same binary becomes a Model Context Protocol server, so an AI agent — Claude, Codex, Cursor, whoever — can drive your Marquee without paying the fork-and-exec cost of shelling out to the CLI on every operation.

That second job came with a problem we didn't appreciate until we watched agents fail at it. A platform like Fibe has a lot of capabilities: provision a Marquee, launch a Playground, connect a repo through Props, start a Genie agent chat, read logs, manage secrets, fund a wallet. Expose every one as a native MCP tool and you hand the agent a menu with hundreds of items — which, counterintuitively, makes the agent worse. So we split the surface into tiers, where "hidden" never means "gone."

A coding agent out of the box is generic. Claude Code, Codex, Cursor, Gemini — they all boot up knowing nothing about you, your repos, your conventions, or yesterday's conversation. On Fibe we call the running instance a Genie, and the platform's whole job is to make that generic Genie yours before a single token is generated.

"Yours" isn't one thing: a provider choice, credentials we have to protect, skill files, files you dragged in, and — the part people underestimate — memory of what happened last time. None of that lives in the container; it lives in an Agent, a small, boring, persistent config record the Genie reads from. This post walks through that record, down to how it reaches the container.

A user clicks "Start" on a Genie. Forty seconds later there's a Claude Code agent running in a container on their Marquee, reachable over TLS at a stable subdomain, talking to their repo through MCP, with their skills and main prompt already staged on disk. If the container later falls over, something notices within a minute and quietly puts it back.

That whole arc — from a config row in Postgres to a self-healing sidecar — is one job, one Redis lock, six steps, and a surprising amount of opinion about what counts as a failure. This post is a walk through the AgentChat deploy pipeline: what we build, in what order, which steps are allowed to fail and retry, and the one design decision that quietly shaped everything else — that a missed health check is not a reason to kill anything.

A Genie is an AI coding agent that lives next to your code. On Fibe you spin one up, it gets its own subdomain, and you talk to it in a pane in the browser. Behind that pane is a Docker Compose project running as a sidecar on your Marquee. The interesting part is that the same machinery runs Gemini, Antigravity, Claude Code, OpenAI Codex, Cursor, and OpenCode — six providers with six different CLIs, six different credential layouts, and six different opinions about how authentication should work — and the runtime that deploys them does not know, or care, which one it just shipped.

That last sentence is the whole design. When you can add a coding agent without touching the deploy pipeline, the reconciler, or the healthcheck, adding the seventh provider stops being a project and becomes an afternoon.