Openai Tts.Bak 2026 01 28T18:01:23+10:30

Text-to-speech via OpenAI Audio Speech API.

This skill appears to be a simple wrapper around OpenAI's TTS API and will send text to api.openai.com using your OPENAI_API_KEY. Before installing: (1) verify the skill source/owner and consider using a dedicated OpenAI key with limited scope/billing; (2) ensure you have curl and jq installed (the script uses jq for JSON encoding but jq is not declared); (3) be aware the registry metadata is inconsistent — it claims no env vars are needed but the script requires OPENAI_API_KEY, so don't assume no secret will be used; (4) if you need the advertised ~/.clawdbot config behavior, confirm the script actually implements it or update the code; (5) if you don't want your API key used, do not install — the key will be sent to OpenAI whenever the script runs. If you want higher assurance, ask the publisher to correct metadata (list OPENAI_API_KEY, curl, jq) and to add explicit checks for jq and clearer handling of the alternate config file.

Type: OpenClaw Skill Name: openai-tts-bak-2026-01-28t18-01-23-10-30 Version: 1.0.0 The skill is designed to perform text-to-speech via the OpenAI API, which aligns with its stated purpose. The `scripts/speak.sh` script properly sanitizes user input for the 'text' argument using `jq -Rs .` before constructing the JSON payload, preventing injection vulnerabilities. Network calls are exclusively directed to the legitimate `https://api.openai.com/v1/audio/speech` endpoint, and the `OPENAI_API_KEY` is handled as expected for authentication. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, prompt injection attempts against the agent, or obfuscation.