Instructions for squidbay/factory, covering you are the engineer seat of this factory, ⛔ preflight — prove access before you say a single word, boot order (every session, before anything else), first contact — when the factory is brand new and standing rules of this seat.
Boot the Coach seat — the factory's surfacer, one of the four seats (Coach, Team Leader, Engineer, Creative Director) — in the Chat room. Invoke only when the human explicitly types /coach or asks to boot the Coach seat. Loads the seat's full boot prompt live from the team's factory repo.
Boot the Team Leader seat — the factory's centre; planner, auditor, journal keeper, one of the four seats (Coach, Team Leader, Engineer, Creative Director) — in the Cowork room. Invoke only when the human explicitly types /team-leader or asks to boot the Team Leader seat. Loads the seat's full boot prompt live from…
Boot the Creative Director seat — the factory's design lane, read-only on code, one of the four seats (Coach, Team Leader, Engineer, Creative Director) — for the Claude Design canvas. Invoke only when the human explicitly types /creative-director or asks to boot the Creative Director. On the Design surface itself, the…
Boot the Engineer seat — the factory's builder, one of the four seats (Coach, Team Leader, Engineer, Creative Director) — in a Claude Code session with the factory repo attached. Invoke only when the human explicitly types /engineer, asks to boot the Engineer, or the repo's automatic CLAUDE.md boot did not fire. Loads…
The factory's training wheels — plain-English coaching for the human who runs the team. Explicit-invoke: run when the human types /factory-coach or asks what something means (repo, branch, PR, merge, deploy, DNS, a seat's role, anything). This skill EXPLAINS and never DOES — one concept at a time, always connected…
Render-and-measure receipts for any HTML page your factory builds — the render half of the design quality gate. Engineer runs it to screenshot every screen size and MEASURE what a source read or a single screenshot only guesses at: horizontal overflow, computed type sizes, tap-target sizes, safe-area presence, mono…
The factory learning to run itself better. Explicit-invoke: run when the human asks for a retro ('how are we doing?', 'what keeps going wrong?', '/factory-retro'), or on a schedule the human sets. Reads the journal end to end, finds friction patterns — repeated misses, slow handoffs, rules that keep tripping…
The Team Leader seat's security read on a PR, run as part of its audit BEFORE the human merges. Explicit-invoke: run when a PR touches credentials, workflows, auth, payments, or personal data — or when the human asks 'is this safe to merge?'. Reads the change in full, reports only high-confidence, real problems in…
Explains and drives the factory's update path — how improvements to the public template reach this office as a human-gated pull request. Invoke when the human asks 'is my factory up to date?', when a factory-update PR needs explaining hunk by hunk, or when an update was declined and needs a selective re-apply.
Mint new custom skills for this factory, properly. Explicit-invoke: run when the human or Team Leader decides a repeated task should become a skill ('make this a skill', 'we keep re-explaining this'). This is a POINTER skill: it walks the seat through fetching and using Anthropic's own skill-creator from…