Agent黄
Community@backtthefuture
Agent黄 publishes skills for social media transcript extraction, SBTI personality card rendering, and multi-expert advisory decision mapping.
Agent Skills by Agent黄
Showing 3 vetted skills indexed across 2 GitHub repositories.
Backtthefuturesocial-sbti
Fetch public JiKe and X posts and render an SBTI personality card to HTML or PNG.
Backtthefuturevideo-transcript
Transcribe videos and podcasts into timestamped transcripts using local FunASR models.
Backtthefutureadvisory-board
Coordinate a multi-phase advisory process with 12 experts to surface risks and produce a decision map.
Frequently Asked Questions About Agent黄
FAQPage SchemaWhat tasks can I accomplish with Agent黄's skills?▼
You can extract verbatim transcripts from Bilibili, Douyin, Xiaohongshu, YouTube, WeChat Channels, Xiaoyuzhou podcasts, and local media; render SBTI personality cards from JiKe and X posts to HTML or PNG; and run a 12-expert advisory process producing a decision map.
Who should use these skills?▼
Content creators, podcast producers, and social media analysts who need timestamped transcripts, speaker-separated interview text, or personality cards from public posts. Decision-makers can also use the advisory-board skill to surface risks across a structured multi-phase expert review.
How does the video-transcript skill run in practice?▼
Paste a supported platform link or provide a local media file. FunASR transcribes on-device using SenseVoice-Small for video and paraformer plus CAM++ for podcasts. Link parsing requires internet; WeChat Channels first use requires scanning a QR code to log into Tencent Yuanbao.
Does transcription require an API key or paid service?▼
No. The video-transcript skill runs FunASR locally on your machine, so no API key is needed for speech recognition. Only link resolution requires network access, and WeChat Channels requires a one-time local QR-code login.
What output formats and dependencies are supported?▼
Outputs include spoken verbatim transcripts, cleaned optimized versions, speaker-blocked podcast text, plain-text PDF, and keyframe screenshot PDF; MP4-only download is available on request. The skill uses Read, Write, Edit, Bash, Glob, and Grep permissions.