Documentation Index
Fetch the complete documentation index at: https://docs.coconut.dev/llms.txt
Use this file to discover all available pages before exploring further.
What you get
One knowledge document —RESEARCH.md, for example — that always reflects the current state of one topic. The skill runs on a schedule (say, every morning at 7am), picks up what’s new, and rewrites the doc. Themes get promoted as they pick up steam, demoted as they cool off, and retired when they fade. The document doesn’t grow over time — it just gets sharper.
Good fit when your topic moves too fast to track by hand, and a one-off report would be stale by the end of the week.
What’s inside
- Skill:
marketing-ai-research(rename it to match your topic) - Job: Runs daily at a time you pick
- Knowledge file:
RESEARCH.md(one file, refreshed every run)
Set it up
Paste this into the Coconut Assistant. Swap the bracketed bits for your topic.
I want a skill that researches [your topic — e.g. the intersection of marketing and AI]. It should look for the latest signals, group them into themes, and only name a theme once it has enough supporting evidence. Over time, it should update the document — not just append to it. Promote themes that are picking up, demote ones that are cooling off, and drop ones that have gone quiet.
Save the document to .nut/knowledge/RESEARCH.md. Each theme should include a short summary, the supporting signals, a note on why it matters for [your audience — e.g. CMOs hosting marketing events], and a strength rating: Emerging → Building → Established → Fading.
Once we’ve agreed on the skill, install it on [coconut name] and set up a daily job to run it at [time, e.g. 7:00am].