365 Skills

Name: 365-skills
Author: Agents365-ai

Production-grade skills for AI coding agents by Agents365-ai. Agent-agnostic, works with Claude Code, Cursor, Copilot, OpenClaw & more.

English | 中文

Install

# Any agent (Claude Code, Cursor, Copilot, etc.) — agent-agnostic
npx skills add Agents365-ai/365-skills -g

# Claude Code plugin marketplace (optional)
/plugin marketplace add Agents365-ai/365-skills

Available plugins

Development & CLI design

Plugin	Description
`agent-native-design`	CLI design for AI agents — evaluate, design, and refactor CLIs to serve humans, agents, and orchestration systems simultaneously
`pi-plugin-cc`	Drive the Pi coding agent from Claude Code — model-agnostic delegation and code review. `/pi:review` with structured findings, `/pi:adversarial-review`, `/pi:rescue`, `/pi:parallel-rescue`, racing/fallback across providers, incremental review

Drawing & diagrams

Plugin	Description
`drawio`	Draw.io diagrams — PNG/SVG/PDF export with visual review loop
`mermaid`	Mermaid diagrams — text-based, GitHub-native, auto-layout
`excalidraw`	Excalidraw — hand-drawn whiteboard style
`plantuml`	PlantUML — UML, C4, sequence, class diagrams
`tldraw`	Tldraw — infinite canvas, sketch-style

Scientific research

Plugin	Description
`semanticscholar`	Semantic Scholar — academic paper search, citation graph, recommendations, BibTeX export
`paper-fetch`	Paper PDF downloader by DOI / title — 7-source fallback (Unpaywall, S2, arXiv, PMC, bioRxiv, publisher, Sci-Hub) with batch mode and idempotent retries
`scholar-deep-research`	End-to-end literature review pipeline — 8-phase script-driven workflow, 7 federated sources (OpenAlex, arXiv, Crossref, PubMed, DBLP, bioRxiv, Exa), cross-source dedup, dual-backend citation chasing, parallel deep-read fan-out, mandatory self-critique, cited reports across 5 archetypes
`asta`	Ai2 Asta MCP — Semantic Scholar academic graph over MCP (no Python). Intent-to-tool routing, safe `fields` defaults, citation traversal, snippet evidence retrieval, and DOI/arXiv/PMID via `externalIds`
`journal-abbrev`	Journal name abbreviation lookup — ISO 4 + MEDLINE, multi-source cascade (JabRef → AbbrevISO → NLM), BibTeX rewrite with `--idempotency-key`, atomic cache rebuild, agent-native JSON envelope with stable error codes and dry-run
`target-prioritization`	Multi-source drug-target due-diligence — turn a ranked gene list (e.g. scRNA-seq DE output) into a per-gene dossier across UniProt, OpenTargets, and PubMed, plus a local cross-lineage DE scan, then re-rank by a configurable composite score (cross-lineage convergence + druggability + disease genetics + tractability + novelty). Disease-agnostic

Knowledge & notes

Plugin	Description
`obsidian-organizer`	Keep a large Obsidian vault tidy — file new notes into the best-fit folder and audit/reorganize existing structure, driven by a single source-of-truth map note (`00_Index/Folder_Map.md`) inside the vault. Link-safe by design (moves go through the `obsidian` CLI so wikilinks auto-repair), propose-then-confirm for bulk changes

Media & creative

Plugin	Description
`ttscn`	Multi-platform Chinese TTS text-to-speech — 14 backends (Edge/Doubao/CosyVoice/Qwen/StepFun/Zhipu/Azure/Tencent/Baidu/MiniMax/Xunfei/ElevenLabs/OpenAI/Google), agent-native CLI with JSON envelope, schema introspection, voice cloning, SSML, emotion, dialects, filterable HTML comparison page
`imagencn`	AI image generation via Alibaba Bailian, ByteDance Volcano Ark & Tencent Hunyuan — 23 models, Chinese text excellence, rich terminal UI, smart config
`videogencn`	AI video clip generation with Chinese video models — text-to-video, image-to-video, first/last-frame and reference-to-video across Bailian (Wan/PixVerse/Kling/Vidu), Jimeng (doubao-seedance), MiniMax (Hailuo), Hunyuan
`assetseeker`	Search free commercial-use creative assets — photos, illustrations, icons, video footage, music, sound effects, and fonts across Pexels, Unsplash, Pixabay, Iconify, Freesound, Google Fonts and more
`video-podcast-maker`	Automated topic-driven video podcast creation — research → script → TTS (7 backends) → 4K Remotion render → BGM mix → Remotion-native subtitles. Multi-platform output (Bilibili / YouTube / Xiaohongshu / Douyin / WeChat Channels), horizontal long-form (16:9 4K) and vertical shorts (9:16), 15-step workflow with mandatory Studio preview
`bangumi-frames`	Bilibili anime frame & character organizer — download a bangumi/UP video (or local file), extract scene-change keyframes, split scenery vs character crops, cluster by CCIP identity or pull one character via a reference folder; optional OCR+LaMa subtitle/watermark removal
`yt2bb`	Repurpose YouTube videos for Bilibili — download via yt-dlp, transcribe with whisper, generate bilingual (English-Chinese) SRT subtitles, hardcode them with ffmpeg

Install individual plugins

Agent-agnostic install: npx skills add Agents365-ai/365-skills and pick the skill. Claude Code plugin path:

/plugin install drawio

Development

Plugin skills are direct copies of the source repos (not submodules). This repo is the central distribution point: update the source skill repo first, then copy the update here and bump the plugin's version in .claude-plugin/marketplace.json:

cp -r ../drawio-skill/skills/drawio-skill/* plugins/drawio/skills/drawio-skill/
# bump "version" for the plugin in .claude-plugin/marketplace.json
git add plugins/drawio && git commit -m "chore: sync drawio-skill"

Most source repos are now private; for those plugins this marketplace is the only distribution channel.