Package Exports

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (codivupload-skills) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

CodivUpload AI Skills

Specialized AI skills for CodivUpload — the social media platform with three first-class interfaces (visual dashboard, REST API, MCP server for AI agents) covering 7+ launched platforms (with Bluesky in active rollout).

These are Markdown-based skills following the Anthropic Skills format. They give Claude Code, Claude Desktop, and other agentic LLMs deep context on how to use CodivUpload effectively for each social media platform.

Available skills

Skill	Best for	Use the skill when…
`codivupload`	Generic catch-all — covers every endpoint	User mentions CodivUpload, wants to publish/schedule social media posts, or needs to call the API
`codivupload-instagram`	Instagram Reels, carousels, feed posts	User mentions Instagram + scheduling, automation, or API
`codivupload-youtube`	YouTube Shorts, long-form, BYOP, 24/7 live streams	User mentions YouTube + bulk upload, BYOP, channel network, or live stream
`codivupload-tiktok`	TikTok Direct Post, Draft, brand content disclosure	User mentions TikTok + automation, AI content disclosure, or branded content
`codivupload-x`	X (Twitter) text/image/video, BYOK rate limits	User mentions X (Twitter) + thread/scheduling/BYOK
`codivupload-facebook-linkedin`	Facebook Pages + LinkedIn personal/Company Pages	User mentions Facebook Page or LinkedIn Company Page automation
`codivupload-agency`	Workspace management, branded invites, whitelabel	User runs an agency, manages multiple clients, or needs whitelabel
`openclaw`	OpenClaw local AI assistant (WhatsApp / Telegram / Discord / Slack)	User runs OpenClaw locally and wants their on-machine agent to publish/schedule via CodivUpload

Installer — picks the right folder for your AI client

# Interactive — asks which client you use
npx codivupload-skills

# Explicit per-client
npx codivupload-skills --client claude              # ~/.claude/skills/ (global)
npx codivupload-skills --client claude-project      # ./.claude/skills/ (project-local)
npx codivupload-skills --client antigravity         # ./.agent/skills/ (Google Antigravity, project-local)
npx codivupload-skills --client antigravity-global  # ~/.gemini/antigravity/skills/
npx codivupload-skills --client cursor              # ./.cursor/rules/ (rule files)
npx codivupload-skills --client zed                 # MCP-only setup printed
npx codivupload-skills --client chatgpt             # MCP-only setup printed

# Specific skill only
npx codivupload-skills --client antigravity --skill instagram

# Custom path
npx codivupload-skills --target /custom/path/

# List available skills
npx codivupload-skills --list

# Help
npx codivupload-skills --help

Per-client install paths

Client	Default folder	Format
Claude Desktop / Claude Code (global)	`~/.claude/skills/`	Anthropic Skills (folder per skill)
Claude Code (project-local)	`./.claude/skills/`	Anthropic Skills — commit to share with team
Google Antigravity (Gemini, project-local)	`./.agent/skills/`	Antigravity Kit (folder per skill, identical to Anthropic Skills)
Google Antigravity (global)	`~/.gemini/antigravity/skills/`	Antigravity Kit
Cursor IDE	`./.cursor/rules/`	Cursor rule files (single .md per skill)
Zed Editor	(MCP-only)	No skills folder; configure MCP server in `~/.config/zed/settings.json`
ChatGPT Pro	(MCP-only)	No skills folder; register MCP Connector at `chat.openai.com/connectors`

After install, restart your AI client to load the skills. The MCP server (codivupload-mcp) is always required — it provides the actual tools.

Manual install (if you prefer)

git clone https://github.com/Codivion/codivupload-skills.git
cp -r codivupload-skills/instagram ~/.claude/skills/codivupload-instagram

How to use with ChatGPT / Cursor / Zed

These skills are written for the Anthropic Skills format, but the underlying integration works via CodivUpload's MCP server. Configure the MCP server in your client and the skill content gives the LLM extra context on best practices.

{
  "mcpServers": {
    "codivupload": {
      "command": "npx",
      "args": ["-y", "codivupload-mcp"],
      "env": {
        "CODIVUPLOAD_API_KEY": "cdv_your_api_key"
      }
    }
  }
}

License

MIT — fork, modify, redistribute. Skills are documentation + prompts, not code that holds secrets.

Distribution

These skills are published in two places:

This repo (canonical source) — github.com/Codivion/codivupload-skills
Anthropic Skills marketplace — when generally available

For ChatGPT users, register the CodivUpload MCP Server as a Connector at chat.openai.com/connectors.