{
"markdown": "\n\n# hs · Huasheng CLI\n\n**From one sentence to a finished, publishable video**\n\n[](https://github.com/superlcr/huasheng-cli/releases/latest)\n[](https://github.com/superlcr/huasheng-cli/releases/latest)\n\n[简体中文](README.zh.md) · **English**\n\n
\n\n---\n\n`hs` brings [Huasheng](https://www.huasheng.cn)'s video creation pipeline to the command line.\nGive it a sentence or a script, and it handles storyboarding, narration, footage and\ncomposition — producing a video you can export or publish directly. You can step in and\nadjust at any point along the way.\n\n**A single self-contained binary.** No Node, no Python, no runtime to install.\nEvery command supports `--json`, designed for scripts and AI clients.\n\n## Step 1: install and sign in\n\nDo this once, regardless of which client you use later.\n\n### Install\n\nPick whichever suits you — all three give you the same binary.\n\n**macOS / Linux**\n\n```bash\ncurl -fsSL https://raw.githubusercontent.com/superlcr/huasheng-cli/main/install.sh | sh\n```\n\n**Windows**, in PowerShell:\n\n```powershell\nirm https://raw.githubusercontent.com/superlcr/huasheng-cli/main/install.ps1 | iex\n```\n\nEither installer downloads the package for your platform, **verifies its SHA256**, and extracts\nit to `~/.local/bin` (`%LOCALAPPDATA%\\Programs\\hs` on Windows).\n\n**With npm**, if you already live in the Node ecosystem or just want to try it first:\n\n```bash\nnpx @superlcr/hs --help\nnpm install -g @superlcr/hs # the command is still `hs`\n```\n\nThe npm package is a small launcher; the binary for your platform arrives as an optional\ndependency, so nothing is downloaded or compiled at install time.\n\n\nManual download instead of the installer
\n\n
\n\nGrab the package for your platform from\n[Releases](https://github.com/superlcr/huasheng-cli/releases/latest), extract it, and place\nthe executable anywhere on your `PATH`:\n\n| Platform | File |\n| :--- | :--- |\n| macOS · Apple Silicon | `hs-darwin-arm64.tar.gz` |\n| macOS · Intel | `hs-darwin-x64.tar.gz` |\n| Linux · x64 | `hs-linux-x64.tar.gz` |\n| Windows · x64 | `hs-windows-x64.zip` |\n\nEvery release ships a `SHA256SUMS` file. **Verifying it is recommended:**\n\n```bash\nshasum -a 256 -c SHA256SUMS\n```\n\n> Both macOS packages are signed and notarized by Apple (Developer ID Application).\n> The Windows package is unsigned; SmartScreen may prompt on first run — choose\n> \"More info → Run anyway\".\n\n \n\n### Sign in\n\nOpen a new terminal, sign in, and confirm that `hs` can read your credit balance:\n\n```bash\nhs auth login\nhs account\n```\n\nThe CLI and every AI client share `~/.hs/credentials.json`; you do not sign in separately.\n\n\n## Step 2: choose how you use it\n\nBoth paths use the same `hs` binary and the same sign-in — pick either, or both.\n\n### Option 1: use hs CLI directly\n\nUse this path for exact commands, scripts, or batch jobs. `hs make` can start from one sentence or\na complete script, run the creation workflow, wait for the finished video, and download it.\n\nCreate a video from one sentence:\n\n```bash\nhs make --script \"Three little-known facts about West Lake\" --out ./out.mp4\n```\n\nRequest an MG-style video:\n\n```bash\nhs make --script \"Explain Song dynasty tea whisking in 30 seconds\" --mode mg --out ./tea.mp4\n```\n\nRead a long script from a file:\n\n```bash\nhs make --script @script.txt --out ./video.mp4\n```\n\n`hs make` approves the storyboard for you — that spends credits, and it prints how many. To read\nthe storyboard and its price first, use the step-by-step commands instead. See the\n[hs CLI guide](docs/cli.md) for parameters, step-by-step editing, resuming, and exporting. See\n[Scripting and automation](docs/automation.md) for JSON, exit codes, and batches.\n\n### Option 2: use hs through MCP in an AI client\n\n`hs` includes an MCP server. Any AI client that supports local STDIO MCP can launch it with:\n\n```json\n{\n \"mcpServers\": {\n \"huasheng\": {\n \"command\": \"hs\",\n \"args\": [\"mcp\", \"serve\"]\n }\n }\n}\n```\n\nThis configuration simply tells the client to run `hs mcp serve` when Huasheng is needed. There is\nno separate hs MCP package to install, and you should not keep the command running yourself. If the\nclient cannot find `hs`, replace `command` with the full path from `which hs` (`where hs` on Windows).\n\nThe following are setup examples for four common clients. For any other MCP client, enter the same\n`command` and `args` in its MCP server settings.\n\n#### ChatGPT Desktop App\n\n1. Open **Settings → MCP servers → Add server**\n2. Enter `huasheng` and choose **STDIO**\n3. Set Command to the full path to `hs`; add `mcp` and `serve` as the two arguments\n4. Save and restart, then type `/mcp` and check that `huasheng` is connected\n\nChatGPT Desktop renders interactive timeline, preview, footage, and export cards. It shares\n`~/.codex/config.toml` with Codex CLI, so this setup also enables hs there.\n\n#### Claude Desktop App\n\n1. Download **[huasheng.mcpb](https://github.com/superlcr/huasheng-cli/releases/latest/download/huasheng.mcpb)**\n2. Double-click it, then select Install in Claude Desktop\n3. Confirm the path to `hs`; the default is `~/.local/bin/hs`\n\nIf you changed the install location, paste the full path from `which hs` (`where hs` on Windows).\nContinue if the first install warns that the extension is unsigned. Claude Desktop also renders\ninteractive cards.\n\n#### Codex CLI\n\n```bash\ncodex mcp add huasheng -- hs mcp serve\ncodex mcp list\n```\n\nDo not add it again if you already configured `huasheng` in ChatGPT Desktop; both read\n`~/.codex/config.toml`. See the [OpenAI MCP documentation](https://developers.openai.com/codex/mcp).\n\n#### Claude Code CLI\n\n```bash\nclaude mcp add --scope user huasheng -- hs mcp serve\nclaude mcp list\n```\n\nBoth commands run in the same terminal where you just signed in, so plain `hs` resolves; if your\nshell cannot find it, substitute the full path from `which hs` (`where hs` on Windows). Codex CLI\nand Claude Code present complete text results instead of desktop interactive cards.\n\n#### Use it through conversation\n\nAfter setup, say in your AI client:\n\n> Make me a 30-second video about why the sky is blue\n\nYou can inspect and refine existing projects too:\n\n> List my recent Huasheng projects\n\n> Make the narration in clip 2 shorter\n\n> Replace clip 3 with more futuristic footage\n\nConfirming a storyboard spends credits, and publishing makes the video public. The client asks\nbefore either action.\n\n## More documentation\n\n- [hs CLI guide](docs/cli.md): step-by-step creation, states, and command groups\n- [Scripting and automation](docs/automation.md): JSON, exit codes, batch control, and IDs\n- [Sign-in, privacy, and requirements](docs/security.md): credentials, network boundaries, platforms\n\n## Safety boundaries\n\n- The CLI and every AI client share one local credential; `hs` never receives your Bilibili password.\n- Approving a storyboard spends credits, and publishing goes public. Both require your confirmation.\n- Scripts and footage are uploaded to Huasheng for video creation; there is no separate telemetry\n channel or background updater.\n\n## Upgrading\n\n```bash\nhs upgrade\n```\n\nThis simply re-runs the installer. `hs` performs **no** update checks and **no** silent\nbackground updates.\n\n## Feedback\n\nFound a problem? Please open an\n[issue](https://github.com/superlcr/huasheng-cli/issues) and include the output of\n`hs --version` — it carries the commit and build time, which is the key to diagnosing anything.\n",
"bytes": 7655,
"sha": "5ff5c62908759603678fe1ab759b56d750c76e5dee786454a6c6487071d93dec",
"repo_slug": "superlcr/huasheng-cli",
"fonte": "repo",
"truncated": false,
"api": "https://agentalog.com/api/listings/mcp_io_github_superlcr_bilibili_huasheng_vid_c3dbc6df/readme"
}