The problem
Metis Labs is a content and digital-marketing agency. Drafting was never the bottleneck. The time went into research per article and keeping every piece consistent with the client's brand, which meant a person opening search tools, the client's site and the brand doc for each brief.
What was built
A content pipeline in n8n. It takes a brief, runs structured research before writing, drafts against the client's brand guidelines, and runs a feedback-and-revision loop rather than emitting one draft and stopping.
Two things it does not do that most content automations do: it does not invent internal links, and it does not write from the model's own idea of the topic. Internal links resolve against the client's actual site, and the research step is grounded in live data rather than a single model call. Output is on-brand per client, not one house voice applied to everyone.
Stack
- Orchestration: n8n (self-hosted), managed through the n8n API
- Research: live keyword and ranking data, plus a per-client index of the client's own site
- Models: accessed through OpenRouter
- Content surfaces: Notion, Google Docs and Sheets
Credentials
Our Apollo.io node is published on npm as
n8n-nodes-apollo (MIT, maintainer arturl95,
14 resources, source public). We hold n8n Verified Creator status.