Tts Responder

Convierte texto a audio OGG usando Piper y envía respuestas habladas por Telegram cuando se activa el modo de voz.

This skill does what it says (synthesizes speech and can send it to Telegram), but check these before installing: - The included script will attempt to POST audio to https://api.telegram.org using environment variables BOT_TOKEN and CHAT_ID. Those variables are sensitive (BOT_TOKEN is equivalent to a secret) and are not declared in the registry metadata—confirm you are willing to provide them and understand the bot's permissions. - Verify the BOT_TOKEN and CHAT_ID values and that you trust the destination chat. A compromised token would allow someone to send messages as your bot. - Piper may auto-download voice models from external hosts the first time it runs. If you need to avoid external downloads, prefetch models in a controlled environment. - The script writes temp files to /tmp and then removes them; you may want to run it in a sandbox or container if you want stronger isolation. - If you plan to integrate this into an agent, confirm the agent will only pass intended messages to the skill—avoid letting it forward arbitrary or sensitive content automatically. If you want to proceed, ask the author or maintainer to update the skill metadata to declare BOT_TOKEN and CHAT_ID (and mark them as sensitive) so the requirements are explicit. If you cannot verify the source or do not want to provide a bot token, do not install.

Type: OpenClaw Skill Name: tts-responder Version: 1.0.0 The skill provides legitimate Text-to-Speech (TTS) functionality by converting text to audio using Piper and FFmpeg, then sending it to Telegram. The shell script 'tts-responder.sh' and the instructions in 'SKILL.md' are transparent, align with the stated purpose, and use standard environment variables for Telegram API authentication without any signs of malicious intent or unauthorized data exfiltration.