fabius-machinalisted

fabius's automation layer — wire deterministic service-to-service workflows and prove the wiring before it runs live. This is no/low-code integration glue (n8n-class): triggers, nodes, webhooks, scheduled jobs, API-to-API connections, branching, and error paths. It is NOT agent orchestration — that's fabius-cohors. The line: machina wires deterministic steps across SaaS/APIs where each step's behavior is fixed; cohors orchestrates LLM agents whose behavior is generative. Use when the user says "automate X", "build a workflow", "connect A to B", "when X happens do Y", "set up a webhook / a cron / an integration", or names n8n / Zapier / Make. The build discipline, the tool-per-intent map, the silent-failure gotcha catalog, and the deployment gate live in references/automation-playbook.md.
shear559/fabius · ★ 2 · AI & Automation · score 64

Install: claude install-skill shear559/fabius

# Fabius Machina — wire the steps, prove the wiring *Machina* — the contrivance, the working mechanism. An automation is a machine made of other people's APIs: it runs unattended, it moves real data, and a silent miswire double-sends or drops items with no error. So the discipline is the same as everywhere in fabius — **scout the live system, strike the smallest correct wiring, prove it before you trust it.** ## 1. Machina vs cohors — pick the right kind of orchestration - **machina** — *deterministic* glue. Fixed steps across SaaS/APIs: a webhook fires → transform → call an API → branch → write. The behavior of each node is known in advance. Reliability and idempotency are the whole game. - **cohors** — *generative* orchestration. LLM agents whose output isn't fixed; the work is least-privilege tools, output contracts, and the agent patterns. A workflow that *contains* an AI step is still machina at the seams (the wiring) and cohors at that one node (the agent). Own each at its layer; don't collapse them. ## 2. The build discipline — never one-shot a live workflow 1. **Discover from the live schema, never from memory.** A platform's node/API surface drifts between releases; read the *current* schema for every node you place. Memory is where silent failures begin. 2. **Build incrementally with surgical edits.** Prefer a par