AmagiLabs
Popular repositories Loading
-
amagi-watch
amagi-watch PublicRuns locally, no API key. Give Claude a video URL and get back a timestamped transcript and key frames, merged into one markdown file.
Python 1
-
amagi-clipper
amagi-clipper PublicTurn a long podcast, webinar or stream VOD into vertical 9:16 clips with burned-in captions. Local faster-whisper + ffmpeg, no API key.
-
amagi-sniff
amagi-sniff PublicFind text hidden from people but readable by AI in PDFs, Word and PowerPoint files, web pages, Markdown and repos: white or tiny text, hidden runs, CSS-hidden HTML, Unicode tag smuggling. No model,…
Python 1
-
-
amagi-screen
amagi-screen PublicLet Claude watch you reproduce a bug: key frames, on-screen text, and a timeline from any screen recording.
Python
-
amagi-listen
amagi-listen PublicLet Claude hear a podcast or meeting recording: a chaptered, timestamped transcript from any local file or URL.
Python
Repositories
- amagi-model-sunset Public
Finds hard-coded LLM model ids in your repo and tells you which ones the provider is retiring, and when
- amagi-benchy-archive Public
Keep every llama-benchy run from every machine in one archive, with GPU watts joined in: leaderboard, tokens per joule, regressions, and one static HTML page.
- amagi-deck Public
Turn a .pptx, .key or slide PDF into one markdown file: a PNG render of every slide plus its text, speaker notes, tables and chart data, so Claude can review a deck it can't open.
- amagi-democast Public
Turn the shell blocks in your README into a demo GIF from a real run, and fail CI when the tool's output no longer matches the GIF. Pillow only: no ffmpeg, ttyd, Go or Node.
- amagi-trace Public
Turn a failed Playwright run (trace.zip or test-results/) into markdown your AI agent can read: steps, failing locator, console errors, failed network calls and the page at failure.
- amagi-cachelens Public
Shows where the Claude Code prompt cache missed, why, and what each miss cost
- awesome-ai-security Public Forked from ottosulin/awesome-ai-security
A collection of awesome resources related AI security
- awesome-prompt-injection Public Forked from Joe-B-Security/awesome-prompt-injection
Learn about a type of vulnerability that specifically targets machine learning models
- amagi-sniff Public
Find text hidden from people but readable by AI in PDFs, Word and PowerPoint files, web pages, Markdown and repos: white or tiny text, hidden runs, CSS-hidden HTML, Unicode tag smuggling. No model, no API key.
- amagi-hookguard Public
Tested Claude Code PreToolUse/PostToolUse hooks: block destructive shell commands, secret-file reads and force pushes to main, keep a local audit log, and check your settings.json hook wiring with `doctor`.
People
This organization has no public members. You must be a member to see who’s a part of this organization.
Top languages
Loading…
Most used topics
Loading…