Skip to content

Other Plugins

Chris Smashe edited this page Jun 17, 2026 · 3 revisions

Action, Post-processing, Memory & Companion Plugins

Beyond transcription engines and LLM providers, TypeWhisper for Linux bundles plugins that run actions, post-process text, store memory, provide companion behavior, and speak text back. This page covers those categories.

The Actions and Memory categories on the Plugins page: Linear, Obsidian, Script Runner, Webhook, File Memory, and OpenAI Vector Memory

Action plugins

Action plugins run custom actions from transcriptions or prompt results.

Plugin Description
Linear Creates a Linear issue from the transcription via a Linear API key, using a configured default team (and optional default project).
Obsidian Saves the transcription as a note into a configured Obsidian vault path and subfolder (default TypeWhisper), either appending to a daily note or writing a new file from a filename template.

Post-processing plugins

Post-processing plugins transform text after transcription.

Plugin Description
Script Runs a user-editable list of scripts through a selectable shell (bash/sh/pwsh, default bash), feeding the text on stdin and using each script's stdout as the new text, with a 5-second per-script timeout.

Memory storage plugins

Memory storage plugins persist and recall extracted memory entries. When one is enabled alongside a configured LLM provider, eligible transcriptions are mined for durable facts that future prompt actions can recall as context — see Memory.

Plugin Description
FileMemory Local JSON storage.
OpenAiVectorMemory Embedding-backed recall.

Companion plugins

Companion plugins observe app/plugin events and react.

Plugin Description
Webhook POSTs (or PUTs) the transcription-completed event to one or more configured webhook URLs — each with its own custom headers and optional profile filter — and keeps a recent delivery log.

TTS providers

TTS provider plugins add spoken-feedback voices that read text back. They implement ITtsProviderPlugin and expose a set of selectable voices.

Plugin Description
Supertonic TTS Local, on-device synthesis. Downloads its model assets on demand and runs with ONNX Runtime — no API key.
OpenAI Cloud TTS via the OpenAI API (the same plugin that provides OpenAI transcription and ChatGPT prompts).
xAI Cloud TTS via the xAI Grok APIs (the same plugin that provides xAI transcription and LLM prompts).

Per-plugin collection settings

The Webhook and Script plugins own user-editable collections and expose per-plugin collection settings under:

  • PluginData/<plugin-id>/

This keeps their entries surviving plugin reinstalls and lets the host edit them through the settings UI. See Plugins for more on the plugin model and where plugin data lives.

Related pages

  • Memory — how memory storage plugins feed long-term recall.
  • Plugins — plugin categories, deployment, and known gaps.

Changelog

Date Change
2026-06-17 Initial version.
2026-06-17 Added Actions / Memory plugins screenshot.
2026-06-17 Replaced one-line plugin descriptions with source-verified behavior (Linear, Obsidian, Script, Webhook); added TTS providers section.

Clone this wiki locally