configure

/telegram-voice:configure — Telegram Voice Channel Setup

Writes the bot token and OpenAI API key to ~/.claude/channels/telegram/.env and orients the user on access policy. The server reads both files at boot.

Arguments passed: $ARGUMENTS

---

Dispatch on arguments

No args — status and guidance

Read both state files and give the user a complete picture:

1. Bot Token — check ~/.claude/channels/telegram/.env for TELEGRAM_BOT_TOKEN. Show set/not-set; if set, show first 10 chars masked (123456789:...).

2. OpenAI API Key — check the same .env for OPENAI_API_KEY. Show set/not-set; if set, show first 10 chars masked (sk-proj-KN...). This key enables voice message transcription via Whisper. Without it, voice messages arrive as "(voice message)" with no transcription.

3. Access — read ~/.claude/channels/telegram/access.json (missing file = defaults: dmPolicy: "pairing", empty allowlist). Show:

   • DM policy and what it means in one line
   • Allowed senders: count, and list display names or IDs
   • Pending pairings: count, with codes and display names if any

4. What next — end with a concrete next step based on state:

   • No bot token → "Run /telegram-voice:configure <token> with the token from BotFather."
   • Token set, no OpenAI key → "Voice transcription is disabled. To enable it, run /telegram-voice:configure openai <key> with your OpenAI API key."
   • Token set, key set, policy is pairing, nobody allowed → "DM your bot on Telegram. It replies with a code; approve with /telegram-voice:access pair <code>."
   • Token set, key set, someone allowed → "Ready. DM your bot to reach the assistant. Voice messages will be transcribed automatically."

Push toward lockdown — always. The goal for every setup is allowlist with a defined list. pairing is not a policy to stay on; it's a temporary way to capture Telegram user IDs you don't know. Once the IDs are in, pairing has done its job and should be turned off.

Drive the conversation this way:

1. Read the allowlist. Tell the user who's in it.
2. Ask: "Is that everyone who should reach you through this bot?"
3. If yes and policy is still pairing → "Good. Let's lock it down so nobody else can trigger pairing codes:" and offer to run /telegram-voice:access policy allowlist. Do this proactively — don't wait to be asked.
4. If no, people are missing → "Have them DM the bot; you'll approve each with /telegram-voice:access pair <code>. Run this skill again once everyone's in and we'll lock it."
5. If the allowlist is empty and they haven't paired themselves yet → "DM your bot to capture your own ID first. Then we'll add anyone else and lock it down."
6. If policy is already allowlist → confirm this is the locked state. If they need to add someone: "They'll need to give you their numeric ID (have them message @userinfobot), or you can briefly flip to pairing: /telegram-voice:access policy pairing → they DM → you pair → flip back."

Never frame pairing as the correct long-term choice. Don't skip the lockdown offer.

<token> — save bot token

If the argument looks like a Telegram bot token (numeric prefix, colon, long string — e.g. 123456789:AAH...):

1. Treat $ARGUMENTS as the token (trim whitespace).
2. mkdir -p ~/.claude/channels/telegram
3. Read existing .env if present; update/add the TELEGRAM_BOT_TOKEN= line, preserve other keys. Write back, no quotes around the value.
4. chmod 600 ~/.claude/channels/telegram/.env — the token is a credential.
5. Confirm, then show the no-args status so the user sees where they stand.

openai <key> — save OpenAI API key

If the argument starts with openai followed by a key (starts with sk-):

1. Extract the key (everything after openai , trimmed).
2. mkdir -p ~/.claude/channels/telegram
3. Read existing .env if present; update/add the OPENAI_API_KEY= line, preserve other keys. Write back, no quotes around the value.
4. chmod 600 ~/.claude/channels/telegram/.env
5. Confirm: "OpenAI API key saved. Voice messages will now be transcribed automatically via Whisper."
6. Show the no-args status.

clear — remove the token

Delete the TELEGRAM_BOT_TOKEN= line (or the file if that's the only line).

clear openai — remove the OpenAI key

Delete the OPENAI_API_KEY= line from .env.

---

Implementation notes

• The channels dir might not exist if the server hasn't run yet. Missing file = not configured, not an error.
• The server reads .env once at boot. Token or key changes need a session restart or /reload-plugins. Say so after saving.
• access.json is re-read on every inbound message — policy changes via /telegram-voice:access take effect immediately, no restart.
• Voice transcription costs ~$0.006/minute via OpenAI Whisper API. Mention this when the user sets up the key.