narrate-video

Generate professional voiceover narration for a video with audio-video sync using Azure TTS. Use this skill whenever the user wants to add narration, voiceov...

What to consider before using/installing this skill: - Credentials: The skill requires AZURE_SPEECH_KEY and AZURE_SPEECH_REGION (it expects them in ~/.narrate_video.env). The registry metadata does not declare these required env vars — ask the author to update the metadata. Only provide a key you trust to be used for TTS and consider using a limited-scope or temporary key that you can revoke. - Inspect the shipped scripts: The included scripts (check_env.py and narration_script_template.py) are readable and show no network exfiltration beyond calls to Azure TTS. Still, review them yourself before running. check_env.py parses ~/.narrate_video.env (it reads values but does not print them); narration_script_template.py loads the env and uses the Azure SDK to send TTS requests. - Local effects: The skill will create files in your project directory (narration_script.py, narration_segments/*.mp3) and temporary frames under /tmp/narration-frames. Ensure you run it in a directory you control and are comfortable with generated files. - Dependencies: You will need ffmpeg, ffprobe, python3, and Python packages azure.cognitiveservices.speech and python-dotenv. Confirm these are installed from trusted sources. - Operational caution: Do not paste your Azure key into a public chat. If unsure, create a temporary key and revoke it after testing. Prefer using secrets managed by your environment rather than leaving credentials in plaintext files in your home directory. - Request fixes from the author: Ask the skill publisher to declare required env vars (AZURE_SPEECH_KEY, AZURE_SPEECH_REGION) in the registry metadata and to document any other required permissions or file-system side effects. Having the metadata accurate will make it easier to audit and approve the skill. Overall: the functionality is coherent and expected for a TTS-based narration tool, but the missing credential declaration in the registry metadata is a meaningful inconsistency — treat the skill as suspicious until the metadata and documentation are corrected and you vet the included scripts.

Type: OpenClaw Skill Name: narrate-video Version: 1.0.1 The 'narrate-video' skill is a well-structured tool for generating video voiceovers using Azure TTS. It demonstrates security awareness by explicitly instructing the agent never to read or display sensitive API keys and includes a dedicated script (scripts/check_env.py) that verifies the existence of credentials in ~/.narrate_video.env without leaking their values. The execution logic in scripts/narration_script_template.py uses safe subprocess calls and standard libraries to handle video processing and speech synthesis.