josoroma
Community@josoroma
josoroma maintains a 99-skill Hermes Agent registry spanning AppLoop Next.js development, ML model serving, creative media generation, and desktop productivity integrations.
Agent Skills by josoroma
Showing 98 vetted skills indexed across 1 GitHub repositories.
hermes-gateway
Normalizes Hermes gateway streaming events with server-only authentication and cancellation handling.
apploop-presentations
Guides development of AppLoop Marp presentation features including slides lifecycle, preview, and style persistence.
yuanbao
Send @mentions, direct messages, and query member info in Yuanbao groups.
presentation-inspect-editor
Debug and modify the AppLoop presentation visual inspector for multi-select, drag, resize, and style persistence.
marp-presentations
Authors and edits Marp Markdown slide decks in AppLoop presentation workspaces.
chat-checkpoints
Implements chat checkpointing with git snapshots, session restore, and edit-resend rollback in AppLoop.
project-runtime
Manages Next.js dev server lifecycle, logs, and readiness checks for generated AppLoop projects.
frontend-design
Define layout hierarchy, responsive behavior, and shadcn composition for generated Next.js app interfaces.
theme-system
Applies and migrates Luma/shadcn theme token sets with preview and rollback support.
generated-app-standards
Enforces formatting, export, component, and schema conventions for generated Next.js app code.
visual-selector
Locate and edit preview elements by selector within bounded visual scopes.
security-review
Reviews generated-project edits for path containment, secret exposure, and isolation violations.
apple-reminders
Manage Apple Reminders from the terminal using the remindctl command-line tool.
imessage
Send and read iMessage and SMS messages via the imsg CLI on macOS.
findmy
Track Apple devices and AirTags via FindMy.app UI automation on macOS.
apple-notes
Create, search, edit, and export Apple Notes via the memo CLI on macOS.
blogwatcher
Monitor blogs and RSS/Atom feeds using the blogwatcher-cli command-line tool.
polymarket
Query Polymarket prediction markets for prices, orderbooks, and price history via public REST APIs.
llm-wiki
Build and maintain an interlinked markdown knowledge base from ingested sources.
arxiv
Search and retrieve arXiv papers via the arXiv REST API and Semantic Scholar.
research-paper-writing
Guides end-to-end ML research paper production from experiment design to conference submission.
xurl
Post, search, and manage X/Twitter content via the official xurl CLI.
jupyter-live-kernel
Execute stateful Python code through a live Jupyter kernel via the hamelnb script.
test-driven-development
Enforces red-green-refactor TDD cycles with failing tests before production code.
Frequently Asked Questions About josoroma
FAQPage SchemaWhat tasks can I accomplish with josoroma's skill registry?▼
You can build AppLoop Next.js projects and Marp presentations, debug code with TDD and root-cause methods, serve and evaluate LLMs via vLLM and llama.cpp, generate images, music, and video with ComfyUI and AudioCraft, and manage GitHub PRs, Office documents, and macOS apps like Reminders and iMessage.
Who is the target audience for these skills?▼
The registry targets developers working with the Hermes Agent runtime, AppLoop frontend engineers building Next.js and shadcn interfaces, ML researchers running inference and benchmarks, and creative coders producing generative art, music, and video across Linux, macOS, and Windows.
How are these skills installed and executed?▼
Skills are authored as SKILL.md files with YAML frontmatter and loaded by the Hermes Agent at runtime. Many declare prerequisite CLI commands such as remindctl, himalaya, or blogwatcher-cli, environment variables like NOTION_API_KEY, or Python dependencies installed via pip before invocation.
Are these skills open source and free to use?▼
Most skills carry the MIT license, including Hermes gateway, debugging, and ML serving skills. The Office document skills (docx, xlsx, pdf, powerpoint) adapted from Anthropic are proprietary with terms defined in their bundled LICENSE.txt files.
What prerequisites do the ML and research skills require?▼
ML skills declare Python dependencies such as torch, transformers, vllm, segment-anything, and audiocraft, typically on Linux or macOS. The research paper pipeline additionally requires semanticscholar, habanero, scipy, and matplotlib, while ComfyUI needs python3 plus a local or cloud ComfyUI backend.