agentleFS
claude-worldclaude-world/claude-world-studio/.claude/skills/studio/SKILL.md

studio

Skill71 starsChanged 25 days agoMITView on GitHub

Claude World Studio CLI - full command reference for all CLI operations

  • Installs packages

15 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use design.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Microsoft AzureSkill

    design

    Design rules and conventions for the API Center Portal. Use when editing detail pages, card components, badges, layout, navigation, sidebars, or headers. Covers badge styling, metadata placement, page navigation patterns, and layout alignment.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 14 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Default: assistant text → stdout, tool calls → stderr (with --verbose)

    line 119
  2. Distinctive:

    When operating Studio via CLI:

    line 184

“Common” is measured: kept by at least 6 in 10 of the 14 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Claude World Studio CLI

Full CLI for Claude World Studio. Maps 1:1 to the REST API + WebSocket.

Install

# Option A: npm global (recommended)
npm install -g @claude-world/studio
studio <command>

# Option B: From source
git clone https://github.com/claude-world/claude-world-studio.git
cd claude-world-studio && npm install
node bin/cli.js <command>
# or: npm link && studio <command>

# Option C: Electron Desktop App
npm run electron:dev            # development
npm run electron:build          # production (.app + .dmg)

MCP Server Setup

# One-command setup (uvx preferred, auto-cached)
npx @claude-world/studio setup-mcp

# Or legacy venv mode
npx @claude-world/studio setup-mcp --venv

# Update installed servers
npx @claude-world/studio setup-mcp --update

Quick Reference

# Server
studio serve                    # Start web UI
studio status                   # Check if running

# Sessions
studio session list
studio session create --title "Research" --workspace /path
studio session get <ID>
studio session rename <ID> --title "New Title"
studio session delete <ID>
studio session messages <ID> --limit 20

# Chat (WebSocket streaming)
studio chat --message "Find trends" --json
studio chat --session <ID> --message "Publish the best"
echo "What's trending?" | studio chat --json
studio interrupt <ID>

# Accounts
studio account list
studio account create --name "Main" --handle "@me" --platform threads --token TOKEN
studio account update <ID> --token NEW_TOKEN
studio account delete <ID>

# Settings
studio settings get
studio settings detect          # Scan system for MCP tools
studio settings apply           # Apply detected values
studio settings set --language en --default-workspace /path

# Publishing (supports all Threads post types)
studio publish --account <ID> --text "Hello!" --score 85
studio publish --account <ID> --text "Vote!" --poll "Option A|Option B"
studio publish --account <ID> --text "Look!" --image "https://url/img.png"
studio publish --account <ID> --text "Watch" --video "https://url/vid.mp4"
studio publish --account <ID> --text "Slides" --carousel URL1 URL2 URL3
studio publish --account <ID> --text "Mood" --gif-id "GIPHY_ID"
studio publish --account <ID> --text "Read" --link-attachment "https://url"
studio publish --account <ID> --text "TL;DR" --text-attachment article.txt
studio publish --account <ID> --text "Spoiler!" --image URL --spoiler-media
studio publish --account <ID> --text "Secret text" --spoiler-text "7:4"
studio publish --account <ID> --text "24hr only" --ghost
studio publish --account <ID> --text "My take" --quote-post-id "12345"
studio publish --account <ID> --text "Post" --reply-control mentioned_only
studio publish --account <ID> --text "Post" --topic-tag "AI" --alt-text "desc"
studio publish --account <ID> --text "Post" --link-comment "https://url"
studio history --limit 10

# Files
studio file list <SESSION_ID> --depth 2
studio file read <SESSION_ID> src/index.ts

Global Flags

FlagEnv VarDefaultDescription
--json—falseJSON output for all commands
--port NSTUDIO_PORT3001Server port
--host HSTUDIO_HOST127.0.0.1Server host

Chat Command

The chat command connects via WebSocket for real-time streaming.

Auto-create session: Omit --session to auto-create. Session ID is printed to stderr.

Input sources (priority order):

  1. --message "text" flag
  2. stdin pipe: echo "text" | studio chat

Output modes:

  • Default: assistant text → stdout, tool calls → stderr (with --verbose)
  • --json: NDJSON — one JSON object per event line

NDJSON event types:

{"type":"assistant_message","content":"Here are the trends..."}
{"type":"tool_use","toolName":"get_trending","toolId":"..."}
{"type":"tool_result","content":"..."}
{"type":"result","success":true,"cost_usd":0.03}

Ctrl+C sends an interrupt to the session and exits with code 130.

Production Pipeline Examples

Headless: Research → Publish

SESSION=$(studio session create --title "Auto Pipeline" --json | jq -r '.id')
studio chat --session $SESSION --message "Find top 3 trending AI topics in Taiwan" --json > trends.jsonl
RESPONSE=$(grep '"type":"assistant_message"' trends.jsonl | tail -1 | jq -r '.content')
studio publish --account acc123 --text "$RESPONSE" --score 80 --json

Freestyle: Full Autopilot

studio chat --message "Freestyle: find trending topics, research the best, \
  write a Threads post, score it, and publish to @mybrand" --json

Scheduled: Daily Content via Cron

# System cron (daily 9 AM)
0 9 * * * studio chat --message "Freestyle pipeline for @mybrand" --json >> ~/studio.log 2>&1

Or use Studio's built-in scheduler: Settings > Scheduled Tasks with per-account targeting and quality gates.

Multi-Account: Different Personas

studio publish --account brand-tech --text "Technical deep-dive..." --score 85
studio publish --account brand-casual --text "Hey did you hear..." --score 75

Settings Keys

CLI FlagAPI KeyDescription
--languagelanguageUI language (en, zh-TW, ja)
--themethemeUI theme (light, dark, system)
--trend-pulse-pythontrendPulseVenvPythontrend-pulse venv python path
--cf-browser-pythoncfBrowserVenvPythoncf-browser venv python path
--notebooklm-pathnotebooklmServerPathNotebookLM server.py path
--cf-browser-urlcfBrowserUrlCF Browser worker URL
--cf-browser-keycfBrowserApiKeyCF Browser API key
--cf-account-idcfAccountIdCloudflare account ID (cf-api mode)
--cf-api-tokencfApiTokenCloudflare API token (cf-api mode)
--default-workspacedefaultWorkspaceDefault workspace path

Instructions for Claude Code

When operating Studio via CLI:

  1. Always use --json for programmatic access
  2. Check status first: studio status --json before other commands
  3. Chat auto-creates sessions — no need to manually create unless reusing
  4. Parse NDJSON line by line for chat output
  5. Use stdin pipe for long messages or multi-line content
  6. Interrupt via CLI rather than killing the process: studio interrupt <ID>

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.