Find the perfect skill
AgentRecall: Persistent Memory for AI Agents
Productivity
Persistent compounding memory for AI agents. 5 default MCP tools: session_start, session_end, remember, recall, check.
AI Brain Setup
Productivity
Set up or upgrade an AI-powered Obsidian vault. Interactive guide to build your vault structure, CLAUDE.md memory, tools, and journaling in one conversation.
Tech News RSS Feed
Productivity
Fetches and summarizes the latest tech headlines from RSS feeds like Hacker News and Lobsters, with per-source filtering support.
Ask Multiple-Choice
Productivity
Presents a multiple-choice question to the user using the AskUserQuestion tool. Useful for interactive decisions.
Daily Work Log
Productivity
Summarize daily Git activity into a concise work log for a date or range, using the current repository and optional configured projects.
Complete Milestone and Prepare Next Version
Productivity
Archives a completed milestone, updates roadmap and requirements, and prepares for the next version.
Taiwan.md Dispatcher
Productivity
Routes the user to the appropriate twmd-* skill based on fuzzy intent, using system vitals and pending tasks.
Control Alexa Devices via CLI
Productivity
Control Amazon Echo devices and smart home via the alexacli command-line tool. Supports text-to-speech, voice commands, and querying device state.
Pre-Mortem Risk Analysis
Productivity
Structured risk analysis for product launches. Categorizes risks as Tigers (real), Paper Tigers (overblown), and Elephants (unspoken), then classifies by urgency.
Beads Project Task Management
Productivity
Use the Beads (bd) CLI for durable project task tracking, blockers, dependencies, and multi-session handoff. Manage ready work, claims, and follow-ups directly from the shell.
Unshelve a Shelved Session
Productivity
Find and resume a previously shelved session via the bookmark_unshelve MCP tool, by keyword search or by browsing the most recent ones.
Tape-opens: Stack-Unwind Session Review
Productivity
Stack-unwind review of a conversation: outputs a tree of completed tasks and open items, sized by measured prompt counts, not feel. Optionally review another session via .jsonl path or session id.