FlowRunner
PricingContact
Theme
Start Free

Evernote

Productivity

Work with Evernote notes and notebooks through a personal developer token. Agents list notebooks, search notes with Evernote's search grammar, read a note's full content, and create notebooks and notes from a flow.

6 actions API key available
Evernote website ↗ Platform Documentation ↗ Capability data verified 2026-07-28
A flagged research email arrives in the team's shared inbox
The agent extracts the substance, the client name, the topic, and the attachments worth keeping
List Notebooks resolves the right project notebook for filing
Find Notes searches the notebook with Evernote's grammar to see whether a note on this topic already exists
Get Note Content pulls the existing note's body when there is one, so the agent appends context instead of fragmenting it
Create Note files the material as a new note in the project notebook, plain text auto-wrapped as ENML
The researcher reviews the week's auto-filed notes and merges, retitles, or refiles what the agent got imperfect

What This Integration Enables

Evernote's value compounds only if things actually get filed, and filing is precisely the work people skip. This connector makes capture and retrieval a flow's job. FlowRunner agents file inbound material into the right notebook, search the base with Evernote's own grammar, and read note content back out so what lives in notes can drive work elsewhere. One honesty note: Evernote's real API is EDAM, a Thrift binary protocol, not REST. This connector is a best-effort convenience layer that Thrift-encodes each call against your Note Store URL with a personal developer token, and it should be verified against your account before production use. - File emails, summaries, and findings as notes in a chosen notebook, plain text auto-wrapped in ENML - Search with Evernote's grammar: intitle:, tag:, notebook:, created: - Read full notes or just the ENML body for extraction workflows - Create notebooks as new projects spin up - Inventory notebooks into external systems for content audits The strategic value runs both directions. Inbound, agents give the knowledge base the one thing it always lacked: a capture process that does not depend on human discipline. Outbound, they give the rest of the stack access to what the base knows, the meeting commitments, the client history, the research trail, which otherwise benefits only whoever wrote it down. FlowRunner sits in the middle as the coordination layer, with a [human-in-the-loop](/concepts/human-in-the-loop) review on bulk writes so the base stays curated rather than merely large. Years of accumulated notes become working memory for agents instead of an archive nobody opens.

Without FlowRunner

Filing happens when there is time The article, the email, the meeting insight all wait in tabs until they are forgotten
Search knows less than it should The knowledge base only contains what someone had the discipline to file
Notes are a destination, not a source Action items written in notes stay in notes, invisible to task trackers

With FlowRunner

Capture is automatic Emails, findings, and summaries file themselves into the right notebook as they arrive
The base grows without willpower The searchable record reflects what happened, not what got manually curated
Notes feed the rest of the stack Agents read note content back out and turn buried action items into tracked work

Use Case Scenarios

The inbox that files itself

A message arrives in [Gmail](/integrations/gmail) matching the research label. The agent extracts the substance, picks the project notebook via the notebook picker, checks Find Notes for an existing note on the topic, and calls Create Note with a clean title and the message body. The client's shifting requirements, scattered across five emails over three weeks, exist as five dated notes in one notebook, findable with a single intitle: search when the project review comes around.

Action items that escape the notebook

Meeting notes get captured in Evernote, and that is usually where the follow-ups die. A nightly agent runs Find Notes for "tag:meeting created:day-7", pulls each body with Get Note Content, and extracts the commitments: who said they would do what. Each one becomes a task in [Google Tasks](/integrations/google-tasks) with a link back to the source note's GUID. The person who writes careful notes stops being the only person who benefits from them.

Auditing a decade of notebooks before the reorg

Before consolidating years of accumulated structure, an agent runs List Notebooks and syncs every notebook name and GUID into [Airtable](/integrations/airtable). For each notebook, Find Notes with a bare query and a result cap samples what actually lives there. The knowledge manager reviews the inventory, marks notebooks to merge or archive, and the agent files a migration plan note per decision with Create Note. The reorg starts from a map instead of from clicking through two hundred notebooks.

Human-in-Loop Highlight

Create Note has a property that makes it deceptively risky in this connector: there is no delete on the other side. The action list reads, searches, and creates, but nothing here removes a note, so every mistake an automated filing flow makes must be cleaned up by hand, note by note, in the Evernote client. A misconfigured migration that files five hundred emails into the wrong notebook, or a loop that creates the same note on every run, turns a curated knowledge base into a junk drawer that a person spends an afternoon shoveling out. So FlowRunner flows treat bulk creation as a proposal: the agent stages the batch, ten sample notes rendered in full, target notebook named, total count stated, and the knowledge owner approves before the writes run. Create Notebook sits behind the same gate, because structure is even harder to walk back than content. Single-note filing from a trusted source runs freely. Anything that writes at volume into a base someone spent years organizing waits for the person who organized it.

Agent processes routinely
Detects exception requiring judgment
Clear match Continues automatically
Ambiguous Routes to human via email
Human decides
Agent resumes with decision

Agent Capabilities

6 actions

Notebooks

2
  • List Notebooks Lists all notebooks in the account with GUID and name via the EDAM listNotebooks operation. The routing table for filing flows and the starting point for structure audits.
  • Create Notebook Creates a new notebook by name and returns its GUID. Used when projects spin up, and gated in bulk because structure mistakes are cleaned up by hand.

Finding and Reading

3
  • Find Notes Searches notes with Evernote's search grammar, optionally scoped to one notebook, returning lightweight GUID and title metadata with a result cap. The dedup check before filing and the discovery step before extraction.
  • Get Note Retrieves a full note by GUID, including title, notebook GUID, and ENML content. The complete pull when a workflow needs everything about a note.
  • Get Note Content Retrieves only a note's ENML body by GUID. The lighter call for extraction loops that process many notes.

Writing

1
  • Create Note Creates a note with a title, body, and optional target notebook GUID, defaulting to the account's default notebook. Accepts full ENML or plain text, which is auto-wrapped in a minimal ENML document. The operation this page's human gate governs at volume.

Frequently Asked Questions

What can FlowRunner do with Evernote?

FlowRunner agents can run List Notebooks, Create Notebook, and Find Notes in Evernote, plus 3 more actions.

Does connecting Evernote to FlowRunner require OAuth?

No. Evernote connects to FlowRunner with an API key, no OAuth flow required.

Can Evernote trigger a FlowRunner workflow automatically?

Evernote doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

Start building with Evernote

$100 in credits. No card required. Connect in minutes.