DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to dictate prompts to an agent by voice with Ovis

Speak into the chat or terminal line, review the text and send it to the agent.

Credentials. If you use the OpenAI transcription option, your own OpenAI key is stored and used by the app, so protect the key and the machine it runs on.

Who it is forA developer who wants to give longer instructions to an agent by speaking, including mixed Korean and English.

The problemTyping long prompts on a phone or while away from the keyboard is slow and error-prone.

The outcomeYou dictate into the chat or terminal line, check the transcribed words and send them to the agent.

What you need first

  • A computer with herdr 0.9.0 or newer, Bun 1.4 or newer and Node 18 or newer (Node is not needed on Windows)
  • Git, curl and tar on Linux or macOS, or Git for Windows on Windows x64
  • A herdr pane running Claude Code, Codex or another supported agent
  • Optional: Tailscale on the computer and your phone for access away from your desk

Steps

  1. Install Ovis with herdr plugin install devswha/ovis and open it in your browser.
  2. Open a Claude Code or Codex pane in the chat view.
  3. Enter your own OpenAI key in Settings if you want that transcription provider, or use the browser's speech recognition.
  4. Dictate into the chat box or the terminal line.
  5. Review the words, then send them to the agent.

Prompt for your agent

Using the devswha/ovis repo, enable voice input in Ovis for <target pane> using the browser speech option and dictate the phrase 'run the test suite'. Confirm the text appears for review before sending, and report which transcription provider was used.

Caveat. The README says audio and polished text go to the configured provider, and browser speech recognition may use the browser vendor's service.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page