Agent Skills by 42ATOM
Showing 21 vetted skills indexed across 1 GitHub repositories.
42atommedia skill（retired）
Retired media skill replaced by native macOS screencapture for screenshots.
42atomthread
List, activate, and read messages from conversation threads with JSON output.
42atomtodo
Manage to-do lists by adding, listing, and completing tasks via the msgcode CLI.
42atomgen skill
Generate images, selfies, text-to-speech audio, and music from command-line arguments.
42atomfile
Read, write, search, move, and copy files using shell commands.
42atommemory
Store, search, and analyze persistent memory entries for AI agents.
42atomplan-files
Store task plans, milestones, and blockers in Markdown files.
42atomcharacter-identity
Identify the current speaker and update sender identity mappings in a workspace CSV.
42atombanana-pro-image-gen
Generate, edit, and describe images via Banana Pro API.
42atomscheduler
Manage recurring msgcode schedules within the current workspace.
42atomGhost MCP
Expose ghost-* desktop automation tools to AI agents via a local bridge.
42atomvision-index
Index detailed vision capabilities and route provider selection for image analysis.
42atompatchright-browser
Drive Patchright and Chrome for web reading, transcription, and browser state diagnostics.
42atomfeishu-send-file
Send workspace files to a bound Feishu chat using feishu_send_file.
42atomlocal-vision-lmstudio
Analyze images locally via an LM Studio vision model with retries.
42atomscreenshot
Capture screenshots locally on macOS using the native screencapture tool.
42atomtwitter-media
Extract tweet text, author, and media from Twitter/X URLs via fxTwitter API.
42atomveo-video
Generate short videos with Google Veo via Gemini-compatible APIs.
42atomreactions
Send emoji reactions to messages via a local HTTP bridge.
42atomscrapling
Extract structured data from JavaScript-rendered web pages using Scrapling.
42atomsubagent
Delegate complex tasks to codex or claude-code subagents with structured subtask cards.