Agent Skills by OpenReason
Showing 81 vetted skills indexed across 1 GitHub repositories.
yuanbao
Automate Yuanbao group member queries and @mention delivery via gateway APIs.
tg-moderator
Moderate Telegram group chats with conversational, language-matched replies.
dogfood
Identify and report web application defects through systematic exploratory QA.
apple-reminders
Manage Apple Reminders from the terminal using remindctl.
macos-computer-use
Automate macOS desktop actions in the background without stealing cursor or focus.
imessage
Send and read iMessages and SMS via the imsg CLI on macOS.
findmy
Automate FindMy UI actions to read Apple device and AirTag locations on macOS.
apple-notes
Manage Apple Notes from terminal using memo CLI on macOS/iCloud.
blogwatcher
Track blog and RSS/Atom feed updates with read/unread management.
polymarket
Aggregates Polymarket's Gamma, CLOB, and Data endpoints into unified read-only queries for markets, prices, and history.
llm-wiki
Automate building and maintaining an interlinked Markdown knowledge base.
arxiv
Search the arXiv API by keyword, author, category, or ID and parse Atom XML results.
research-paper-writing
Automate ML research paper workflows for NeurIPS, ICML, ICLR, and ACL submissions.
minecraft-modpack-server
Automate modded Minecraft server setup from server packs.
pokemon-player
Automate Pokemon gameplay through a headless emulator with RAM reads.
xurl
Automate posting, reading, and messaging on X via the xurl CLI.
jupyter-live-kernel
Provide a stateful Python REPL via a live Jupyter kernel.
test-driven-development
Guide developers through writing failing tests before implementing minimal code.
systematic-debugging
Identify and resolve software bugs through a structured, phase-driven debugging process.
plan
Generate a timestamped Markdown plan file under .cogitum/plans/.
python-debugpy
Debug Python apps interactively with pdb REPL and remote debugpy (DAP).
writing-plans
Generate structured implementation plans with tasks, file paths, and tests.
spike
Run throwaway experiments and produce a README with a formal verdict.
requesting-code-review
Run static security scans, tests, linting, and independent review on staged git changes.