Roxels/ docs
mcp

Claude Code skill: /run-roxels

/run-roxels is a Claude Code skill that turns a Roxels voice call into instructions for your coding session. You type /run-roxels, a conversation opens in your browser, you talk (and optionally screen-share) to dictate what you want, and when you hang up, Claude reads the transcript, plays back the instructions it heard, and acts on them.

Under the hood it runs the conversation against a default template curated by Roxels, then polls the results — you don't configure any template.

Install

Paste this into Claude Code (or send it to a teammate — it's also hosted at /skills/run-roxels/install.txt):

Install the Roxels "/run-roxels" skill for Claude Code:
 
1. Create the folder ~/.claude/skills/run-roxels
2. Download these files into it, keeping the names:
   - https://app.roxels.ai/skills/run-roxels/SKILL.md
   - https://app.roxels.ai/skills/run-roxels/run_interview.py
   - https://app.roxels.ai/skills/run-roxels/README.md
   - https://app.roxels.ai/skills/run-roxels/env.example   (save it as .env.example)
3. Run python3 ~/.claude/skills/run-roxels/run_interview.py once — it prints
   how to finish setup (a .env with my Roxels API key from
   https://app.roxels.ai -> Settings -> API Keys). Write the .env for me.
4. Done — from now on I can type /run-roxels.

The script is plain Python (standard library only) — no virtualenv, no dependencies.

What you need

  • A Roxels API key — generate one at app.roxels.ai → Settings → API Keys. It goes in a .env file next to the script; the skill guides you through this on first run.
  • That's it. Conversations and results stay in your organization.

How results come back

When the call ends, the skill prints the full conversation transcript into your Claude Code session. Claude distills your words into a single instructions prompt, echoes it back ("Your instructions from the call: …"), and executes it — exactly as if you had typed it. The full structured payload (extracted data, statuses, recording link) is saved as JSON next to where you ran it.

Last reviewed 2026-07-08.