ClaudeCodeMod

All shelves / Mods / Workflow

bettercallgpt

By insta-fusion

bettercallgpt is a Claude Code workflow mod by insta-fusion.

Puts this session on a voice call you talk on while Claude codes, and sends microphone audio and call text to Azure or OpenAI on your own key.

Draws

Install

The marketplace name is bettercallgpt.

In Claude Code
/plugin marketplace add insta-fusion/bettercallgpt
/plugin install bettercallgpt@bettercallgpt
Open the repo

It runs inside the Claude Code process. The side panel is only what the author's README names. Read the repo before you install, or run claude plugin validate on the plugin folder.

Version
0.2.2
License
MIT
Repo stars
8
Last change
2026-10-09

What it can reach

  • Draws in the session
  • Reaches the network or an outside service
  • Sees tool calls
  • Runs a command when you press

The voice opens in Chinese, then answers in whatever language you just spoke. macOS only, in the terminal or the desktop app's Code tab, and it needs a Mac microphone. Install uv. Paste the voice key into ~/.config/bettercallgpt/.env, not into the chat. Press Call, Mute, Steer, or Hang up above the prompt, or type /call, /mute, /steer, or /hangup. Call starts the voice process from this repository as a child, or through uvx from the pinned release. The mod itself sends nothing. The voice process sends microphone audio, Claude's reply text, anything you type during the call, your speech, and a one-line permission summary to Azure Voice Live on a cognitiveservices.azure.com host, the default, Azure GPT-Live on an openai.azure.com host, or OpenAI Realtime at api.openai.com, with your own key and on your own account with that provider. Tool output and files are not forwarded. A spoken yes never approves anything. The README says not to allow-list the start, because it opens the microphone and a paid connection. The voice service bills your own account. The tagged release is fetched from GitHub through uv on first use. Earlier transcript text is read on this machine and is not sent.

The install name is bettercallgpt. The marketplace name is bettercallgpt. The marketplace source is ./plugin. The voice call is why this page files it under workflow. wolfbud, a voice coworker you call beside the session, hears what Claude does and sends a prompt you agree on. Your voice and that activity go to ElevenLabs. This call is full duplex on your own Azure or OpenAI key, and a spoken yes never approves. The row is Call, Mute, Steer, and Hang up. Mute makes the voice service hear silence until Unmute. GPT in the name is the voice model, and the README says the project is not affiliated with OpenAI, Microsoft, or Anthropic.

Demo

From the author's repo.

What it hooks into

  • classic.PermissionRequestA Claude Code event
  • session.startWhen a session starts
  • session.endWhen a session ends
  • command.runWhen you run its slash command/call/call-icons/hangup/steer/mute
  • turn.startWhen Claude starts a turn
  • turn.completeWhen a turn ends
  • prompt.submitWhen you send a prompt
  • ui.renderDraws into the sessionAbovePrompt

Read from the on() calls in its hook modules. Open its source under Files to check what each one does.

Slash commands 5

  • /call-iconsSymbols on the call controls: keys, nerd (Nerd Font), emoji or none
  • /callStart a voice call in this session
  • /hangupEnd the voice call
  • /muteMute or unmute your microphone on the voice call
  • /steerHave Claude take what you just said now

Files

The author's README and source, 4 files. They load when you open them.

Facts

Author
insta-fusion
Category
Workflow
Marketplace
bettercallgpt
Repo
insta-fusion/bettercallgpt
Folder
plugin
Version
0.2.2
License
MIT
Repo stars
8
Last change
2026-10-09
Language
Python
Files
10

Other mods in Workflow

allowlist-coachice-lfernandesCounts permission dialogs you answer and, after repeated approvals, offers that rule for the allow list, asking you before it writes.test-hudice-lfernandesDraws each test run on the status line, passing over total with a sparkline of failures, and a pane of the tests that failed.run-lot-panepierrebelinFor cctoolkit, which builds a feature batch by batch, draws a live pane of the batch or the plan, with agents, tools, and tokens.html-shelfronanworksDraws paths in a reply as links that open in the default app, and code blocks as cards you can copy or place in the prompt.actual-cliactual-softwareFor Actual AI, which keeps agents on a team's architecture decisions, draws a row of how many ADRs reached context and adds a file's rules beside each read.autodev-coredjnsty23Holds a commandFor Claude Auto-Dev, an autonomous workflow from brainstorm to ship, pins a status line under the prompt with what its hooks did and the sprint's pass states.kgai-modkgaidevDraws a status line when a decision is recorded, and Record and Skip in confirm mode, for kgai, the team's shared memory of why the code changed.glanceflowAntreas-StrbHolds a commandDraws a checklist above the prompt with the plan, the step Claude is on, how far along it is, and when Claude needs you.batonashishsk93Draws a band above the prompt of this session, the other sessions, and the tasks passed between them.todosbengousDraws the repository's TODOs in a scrolling band above the prompt, newest first, including findings an agent recorded.zcode-executorKyoMioFor zcode-executor, which hands Claude's tasks to the local ZCode agent, draws a live pane of this project's ZCode sessions and the requests waiting on you.arbeitslistenikisgeDraws a to-do field beside the chat, queues tasks while Claude works, and starts the next one, with Claude's plan above it.
All 396 in Workflow