MCP App StoreMCP App Store
from bumblebee16
Assembles fragmovie-style videos by splicing real movie/TV audio clips that match an input phrase; local transcription (faster-whisper) and FFmpeg used — no TTS
Bumblebee takes English phrases and builds a montage video in which each word is sourced from real movie or TV clips. It searches indexed clip sites (yarn.co, playphrase.me fallback), downloads candidate MP4s, transcribes them locally with faster-whisper to get word-level timestamps, cuts clips precisely with FFmpeg, and concatenates them with short fades and pauses. It supports generating multiple variants that avoid clip reuse. The tool intentionally never fabricates audio (no TTS); when words are unobtainable it suggests context-appropriate synonyms and re-runs.
When the user requests a "fragmovie" or to "splice a line from movies", or when they want a montage constructed from authentic spoken-word clips. Not suitable for generating new speech or for non-English phrases.
bumblebee.py and modular Python components (yarn search, downloader, transcriber, cutter, concat).Intended for local execution in Python 3.9+ environments with FFmpeg. Integrates with agent-host CLIs that can run local Python/FFmpeg workflows.
Bumblebee assembles fragmovie-style videos by splicing real movie/TV audio clips matching an input phrase using yarn.co and playphrase.me. SKILL.md is well-structured with clear triggers, architecture overview, and an intelligent synonym substitution loop for unreachable words. No scripts were available to run (repo not cloned). Security concerns include curl_cffi Chrome TLS impersonation to bypass Cloudflare protections and mp4 downloads without checksum verification.
No scripts in DB to run-test; static analysis only. The TLS impersonation to bypass Cloudflare is the main security concern — not malicious but ethically questionable. Skill is well-documented with a clever synonym substitution pattern for handling missing words.
Cost Tracker
Monitor agent session costs, set budget alerts, and get actionable token-spend optimizations to keep multi-session workflows within budget.
Audit Agents/Skills/Commands
Run a quantitative quality audit on agents, skills, and commands to check production readiness and get prioritized recommendations.
NotebookLM Research Agent
End-to-end research-to-content pipeline: ingest sources into Google NotebookLM, run deep cited research, generate downloadable artifacts, and produce polished c
Forcemanager Automation (Rube MCP)
Automate Forcemanager tasks via Rube MCP (Composio); discovers tools, ensures active connections, and runs toolkit workflows safely.
draw.io Diagram Skill
Create, edit, lint, and export draw.io diagrams; includes export helpers, SVG linting, and repository QA for clean architecture and icon-based diagrams.
GitHub Explorer
Deep-dive, multi-source analysis of GitHub projects—architecture, community health, issues, commits, and competitive landscape.
API Development
Guidelines and patterns for building robust REST APIs: correct HTTP status codes, RFC 7807 error responses, validation, sanitization, rate limiting, and central
Harness Health Audit
Assess and score the overall health of a Claude Code harness across architecture, skills coverage, hooks, rules, MCP servers, eval pipelines, and team setup.