Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/LilMGenius/polysona/llms.txt

Use this file to discover all available pages before exploring further.

The $trend / /trend command dispatches the trendsetter agent to scan what is currently trending across your persona’s domains and return a ranked, platform-tagged topic list. Run it at the start of a content cycle — after $introduce — to get a prioritised set of topics before drafting. Results are saved to disk so downstream skills like $content and $qa can reference them.

​
Invocation

$trend

​
What it does

  • Reads personas/_active.md to resolve the active persona and loads personas/{active}/persona.md.
  • Triggers the trendsetter agent to scan current trends matched to the persona’s domains.
  • Returns a ranked topic list with platform-fit tags for each entry:
    • x — X (Twitter)
    • threads — Threads
    • linkedin — LinkedIn
    • naver-blog — Naver Blog
    • brunch — Brunch
  • If live search is slow or unavailable, the trendsetter immediately falls back to persona-driven topic ranking from local files and still returns usable results — it will never stall waiting for an external signal.
  • Saves the ranked scan to content/trends/ before returning its response.
  • Re-reads the saved trend file and includes the confirmed saved path in the response.
  • If no file was actually written, the trendsetter reports failure rather than claiming success.

​
Prerequisites

An active persona must exist before running a trend scan: 
personas/{active}/persona.md    ← must exist (from $interview)
Run $interview first to build a persona, then optionally $introduce to activate it, before calling $trend.

​
Output

The trendsetter saves its ranked scan to: 
content/trends/
The response includes the confirmed file path and a ranked list in the form of: 
1. [Topic title] — platforms: x, linkedin
2. [Topic title] — platforms: threads, brunch
3. ...
Each entry is matched to the persona’s domain expertise and ranked by relevance and current momentum.

​
Guardrails & Notes

The trendsetter must write a file before it reports success. If you receive a response without a confirmed content/trends/ path, the save failed — re-run the command.
  • The fallback to local persona-driven ranking is automatic and silent only when live search is genuinely unavailable. The response will note when a fallback was used.
  • Platform tags reflect persona-domain fit, not general platform popularity. A topic tagged only linkedin is not necessarily unfit for x — it means the trendsetter judged the fit strongest for LinkedIn given your persona.
  • Topic rankings are scoped to the active persona’s domains. Switching the active persona and re-running $trend will produce a different ranked list.
The $trend skill is executed by the trendsetter agent. See the Trendsetter agent page for details on its domain scanning, topic prioritisation, and platform-fit filtering logic.

Build docs developers (and LLMs) love

Get started for freeTalk to us