{"id":47998959,"url":"https://github.com/anyt-io/notebook","last_synced_at":"2026-04-04T12:10:45.422Z","repository":{"id":336728781,"uuid":"1149901390","full_name":"anyt-io/notebook","owner":"anyt-io","description":"AnyT notebook","archived":false,"fork":false,"pushed_at":"2026-03-13T20:29:41.000Z","size":10653,"stargazers_count":9,"open_issues_count":1,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-03-14T04:28:51.587Z","etag":null,"topics":["ai-agents","claude-code","codex","notebook"],"latest_commit_sha":null,"homepage":"https://docs.anyt.io","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/anyt-io.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-02-04T16:48:12.000Z","updated_at":"2026-03-13T20:29:47.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/anyt-io/notebook","commit_stats":null,"previous_names":["anyt-io/notebook"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/anyt-io/notebook","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anyt-io%2Fnotebook","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anyt-io%2Fnotebook/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anyt-io%2Fnotebook/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anyt-io%2Fnotebook/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/anyt-io","download_url":"https://codeload.github.com/anyt-io/notebook/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anyt-io%2Fnotebook/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31398902,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-04T10:20:44.708Z","status":"ssl_error","status_checked_at":"2026-04-04T10:20:06.846Z","response_time":60,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-agents","claude-code","codex","notebook"],"created_at":"2026-04-04T12:10:44.986Z","updated_at":"2026-04-04T12:10:45.408Z","avatar_url":"https://github.com/anyt-io.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# AnyT Notebook\n\n**The workflow development environment for AI agents.**\n\nAI agents are powerful, but they work like chatbots -- you give them a prompt, hope they get everything right in one shot, and when things break halfway through, you start over. AnyT Notebook changes this. It gives AI agents a **workflow layer**: visible steps, human checkpoints, and recoverable state.\n\n![AnyT Notebook Demo](media/demo.gif)\n\n## Why AnyT Notebook\n\nWhen you ask an AI agent to do something complex, the one-shot chatbot model falls apart:\n\n- **No visibility** -- You can't see the plan before it runs\n- **No control** -- You can't pause, review, or redirect mid-execution\n- **No recovery** -- If step 4 of 8 fails, you redo everything\n- **No separation** -- AI handles things a simple shell command could do faster\n\nAnyT Notebook solves this with a Jupyter-style notebook for AI agent workflows. Break tasks into cells, add human checkpoints, mix AI and shell scripts, and run with full control. Edit cells with a rich text editor featuring code blocks with syntax highlighting, or toggle to raw markdown. Give cells human-friendly labels so your notebook reads like a plan. Configure agent profiles per-notebook or per-cell to use different runtimes and settings where they matter most. Enable **session mode** to maintain Claude context across task cells, browse generated files in the **Files tab**, and get Jupyter-style **shell error display** with line highlighting when scripts fail.\n\n## Supported Runtimes\n\nAnyT Notebook works with multiple AI runtimes:\n\n| Runtime | Status | Notes |\n|---------|--------|-------|\n| **Claude Code** | Primary | Full support with streaming output |\n| **Codex** | Supported | Full support with JSON output |\n| **Gemini CLI** | Supported | Full support with streaming output |\n\nSelect your runtime from the dropdown in the notebook toolbar. Switch between runtimes at any time. Configure different runtimes per-notebook or per-cell via VS Code settings -- for example, use Claude Code for complex reasoning tasks and a faster runtime for simpler ones.\n\n## Installation\n\n### From VS Code Marketplace\n\n1. Open VS Code or Cursor\n2. Go to Extensions (Ctrl+Shift+X / Cmd+Shift+X)\n3. Search for \"AnyT Notebook\"\n4. Click Install\n\n### From VSIX\n\n1. Download the latest `.vsix` file from the releases\n2. In VS Code, open Command Palette (Ctrl+Shift+P / Cmd+Shift+P)\n3. Run \"Extensions: Install from VSIX...\"\n4. Select the downloaded file\n\n## Quick Start\n\n**1. Try a Sample** -- `Cmd+Shift+P` -\u003e \"AnyT: Open Sample Notebook\" -\u003e pick one -\u003e click Run\n\n**2. Or Create Your Own:**\n- `Cmd+Shift+P` -\u003e \"AnyT: New Notebook\"\n- Click **+ Task**, **+ Shell**, **+ Input**, **+ Break**, or **+ Note** to add cells\n- Edit cells with the rich text editor (or switch to the Markdown tab for raw editing)\n- Click **Run** to execute the notebook step by step\n\n---\n\n## Cell Types\n\nAnyT Notebook provides five cell types. Mix and match them to build workflows that combine AI execution, automation, and human oversight.\n\n### Task\n\nTask cells are executed by your AI agent (Claude Code, Codex, or Gemini). Write a natural language instruction and the agent carries it out. Use tasks for anything that requires reasoning, code generation, or multi-file changes.\n\n\u003e *Example: \"Create a REST API with Express.js and TypeScript, including user authentication routes.\"*\n\n### Shell\n\nShell cells run scripts directly -- no AI involved. They're fast, deterministic, and ideal for commands you already know. Use them for installs, builds, tests, linting, or any CLI operation. When a script fails, a Jupyter-style error display highlights the exact failing line.\n\n\u003e *Example: `npm install \u0026\u0026 npm run build \u0026\u0026 npm test`*\n\n### Input\n\nInput cells pause execution and present a form to the user. Use them to collect configuration values, choose between options, or supply information that downstream cells need. Form responses are available to subsequent cells.\n\n\u003e *Example: \"Choose deployment target\" with a dropdown for staging vs. production.*\n\n### Note\n\nNote cells are markdown documentation checkpoints. They auto-complete instantly when reached and serve as section headers, context annotations, or progress markers in your workflow.\n\n\u003e *Example: A section header like \"Phase 2: Testing and Validation\" to organize your notebook.*\n\n### Break\n\nBreak cells pause execution so you can review what's happened so far. Inspect outputs, verify results, then click Continue when ready. Add breaks liberally while developing a workflow; remove them once you're confident it runs end-to-end.\n\n\u003e *Example: \"Review the generated project structure before adding API routes.\"*\n\n---\n\n## Workflow Development Lifecycle\n\nAnyT Notebook treats AI workflows like code -- build, test, and refine them iteratively:\n\n1. **Create** -- Add cells visually: tasks for AI work, shell for scripts, inputs for user decisions, notes for documentation\n2. **Debug** -- Insert break cells as checkpoints and run step by step to inspect each output\n3. **Iterate** -- Review results, fix failing steps, reorder or add cells as needed\n4. **Harden** -- Remove breakpoints once confident and run the full workflow end-to-end\n5. **Share** -- Version the `.anyt` file and share with your team\n\nThis is the same cycle as software development: write code, add breakpoints, step through, fix bugs, remove breakpoints, deploy. **AnyT Notebook brings this discipline to AI agent workflows.**\n\n---\n\n## Sample Notebooks\n\n| Sample | What it does |\n|--------|--------------|\n| **Stock Analysis** | Analyze stock data with AI-powered insights |\n| **YouTube Video Summarizer** | Summarize YouTube videos from transcripts |\n| **Story to Video** | Generate videos from story prompts |\n| **LinkedIn Outreach** | AI-powered LinkedIn outreach with agent-browser automation |\n| **Image Background Remover** | Remove image backgrounds using AI |\n| **Gemini Image Playground** | Generate and edit images with Google Gemini |\n| **Page to RedNote** | Convert web pages to RedNote-style carousel posts |\n| **Podcast to Ebook** | Convert YouTube podcasts into formatted ebooks |\n\nOpen Command Palette -\u003e **\"AnyT: Open Sample Notebook\"** -\u003e pick one -\u003e click Run.\n\n---\n\n## Configuration\n\n| Setting | Default | Description |\n|---------|---------|-------------|\n| `anyt.agent` | `\"claude\"` | AI agent (`claude`, `codex`, or `gemini`) |\n| `anyt.autoVersion` | `true` | Auto-version on changes |\n| `anyt.shell.path` | `\"\"` | Custom shell path for shell cells |\n| `anyt.shell.loginShell` | `true` | Use login shell to source profile files |\n| `anyt.claude.sessionMode` | `false` | Maintain Claude context across task cells |\n| `anyt.claude.resumeFailRecovery` | `true` | Auto-retry without session args when resume fails |\n\n---\n\n## Commands\n\n| Command | What it does |\n|---------|--------------|\n| `AnyT: New Notebook` | Create a fresh `.anyt` file |\n| `AnyT: Open Sample Notebook` | Browse and open built-in sample notebooks |\n| `AnyT: Settings` | Toggle extension settings via QuickPick |\n| `AnyT: Toggle Session Mode` | Toggle Claude session mode on/off |\n\nAll other actions (run, stop, reset, add cells) are available directly from the notebook toolbar.\n\n---\n\n## Requirements\n\n- **Claude Code CLI** installed and authenticated ([install guide](https://docs.anthropic.com/en/docs/claude-code)), **Codex CLI** installed, or **Gemini CLI** installed ([install guide](https://github.com/google-gemini/gemini-cli))\n- VS Code 1.85+\n\n---\n\n## Skills\n\nAnyT Notebook supports [PSPM](https://github.com/anyt-io/pspm-cli) skills — reusable prompt skill packages that extend AI coding agents with additional capabilities. This repo includes example skills in the [`skills/`](skills/) directory:\n\n| Skill | Description |\n|-------|-------------|\n| [youtube-downloader](skills/youtube-downloader/) | Download YouTube videos and transcripts in various formats and qualities |\n| [pspm-skill-creator](skills/pspm-skill-creator/) | Scaffold, validate, and package PSPM skills with isolated runtime environments |\n| [create-anyt-notebook](skills/create-anyt-notebook/) | Create new AnyT Notebook files from templates |\n| [remove-background](skills/remove-background/) | Remove image backgrounds using AI |\n| [upscale-image](skills/upscale-image/) | Upscale images using AI super-resolution |\n| [rednote](skills/rednote/) | Generate RedNote-style carousel posts from web content |\n| [ebook](skills/ebook/) | Convert podcast transcripts to formatted ebooks |\n| [gemini-image](skills/gemini-image/) | Generate and edit images with Google Gemini |\n| [gemini-video](skills/gemini-video/) | Generate videos with Google Gemini and concatenation support |\n| [keling-video](skills/keling-video/) | Generate videos with Kling AI and keyframe control |\n\nInstall a skill with PSPM:\n\n```bash\npspm install youtube-downloader\n```\n\nSee [docs/skill-development-guide.md](docs/skill-development-guide.md) for full conventions on creating new skills.\n\n---\n\n## Documentation\n\n- [Product Overview](docs/PRODUCT.md) -- Detailed introduction to AnyT Notebook\n- [User Guide](docs/USER_GUIDE.md) -- Complete guide to using AnyT Notebook\n- [Full Documentation](https://docs.anyt.io) -- Comprehensive documentation site\n- [File Format Specification](docs/anyt-notebook-spec.md) -- Complete `.anyt` format reference\n- [Changelog](CHANGELOG.md) -- Release history and changes\n- [Examples](examples/) -- Sample notebooks to get you started\n\n## Links\n\n- [VS Code Marketplace](https://marketplace.visualstudio.com/items?itemName=AnyT.anyt-notebook)\n- [GitHub Issues](https://github.com/anyt-io/notebook/issues)\n\n## License\n\nCopyright (c) 2026 AnyTransformer Inc. All rights reserved.\n\nSee the [LICENSE](LICENSE) file for details.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanyt-io%2Fnotebook","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fanyt-io%2Fnotebook","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanyt-io%2Fnotebook/lists"}