video-transcript

Use when video content needs to be extracted as text: pasted YouTube links or IDs, requests to transcribe, summarize, quote, translate, convert video to text...

This skill appears to be what it says: it will use a TranscriptAPI key to fetch YouTube transcripts and can create an account for you if you ask it to. Before installing/use: (1) Decide whether you want to paste an API key you create yourself (safer) or let the agent create an account (it will need your email and the OTP). (2) Understand that the skill will ask to store the key persistently (may modify shell profile or agent secret storage) so future sessions can use it — confirm where it will be saved and that this location is acceptable. (3) The agent will instruct writing temporary files for tokens; ensure temporary files are cleaned up and not stored in a shared location. (4) Verify transcriptapi.com and the account’s billing/usage limits and do not reuse high-privilege or unrelated credentials. If you are uncomfortable with the agent editing profile files or handling OTPs on your behalf, prefer to generate the key yourself and paste it when prompted.

Type: OpenClaw Skill Name: video-transcript Version: 1.5.0 The skill includes instructions in 'references/auth-setup.md' that explicitly direct the AI agent to circumvent platform-level security features, specifically by bypassing the redaction of sensitive tokens (API keys) through the use of temporary files. It also instructs the agent to autonomously identify and modify system persistence files (like shell profiles) to store credentials. While these actions are framed as functional setup steps for the 'transcriptapi.com' service, the methodology for evading security filters and modifying environment configurations represents a high-risk pattern often used to hide activity or maintain persistence.