A fixed roster,
not a box of prompts.

Fenlio’s agents are a small, fixed team — each with one job, built and tested individually. Onboarding activates the subset that fits how you work. Nothing here improvises its own role, which is exactly why you can trust what it does to your notes.

The difference is who initiates

A chat box does nothing until you ask. Fenlio’s agents run on your workspace’s schedule, so the answer to “is my brain organized?” is always “yes, as of the last pass” — not “it will be, once you catch up.”

A chat box over your notes
  • You write a prompt every time you want anything.
  • Notes you never ask about are never looked at.
  • Organizing is still your job — the chat just searches the mess.
Fenlio’s agents
  • Every new note is read, filed, and checked without a prompt.
  • Contradictions are flagged before you would have found them.
  • Studio drafts are waiting when you arrive, sources cited.

The roster

This is the complete list. Your onboarding answers decide which of these run in your workspace, and which gets priority.

Ingestion Classifier

Categorizes and files incoming notes into your custom folder architecture.
Runs when new notes arrive

Sync Agent

Monitors update cycles and maintains structural coherence across folders.
Runs on your workspace's sync schedule

Conflict Detector

Detects duplicate assertions and flags contradictory statements across notes.
Runs on every pass

Voice & Brand Watcher

Ensures newly ingested material aligns with your tone and voice notes.
Runs as material is ingested

Recall & Retrieval Agent

Surfaces relevant past notes, frameworks, and client history automatically.
Feeds Studio and the interview

Ideation Agent

Generates topic prompts and content angles based on your stored knowledge.
Feeds Studio's ideation lane

Collaboration Agent

Prepares notes and summaries formatted for team handoffs.
Reserved for team workspaces

The roster is deliberately fixed. Agents that invent their own jobs make errors faster, not work better — so Fenlio’s agents each do one thing, and adding a new one is a product decision we make and test, not something a model decides at runtime.

Put the roster on your notes.

Two questions pick your team. Your first pass runs within minutes of your first upload.