{ "markdown": "\n

Short Video Agent Kit

\n\n
\n \"Short\n
\n\n

\n One agent-first CLI + MCP for short-form AI video.
Sora ยท Veo ยท xAI ยท Seedance — dry-run by default, paid generation when explicitly enabled.\n

\n\n

\n \"npm\n \"npm\n \"License\n \"Built\n

\n\n

\n \"GitHub\n \"CI\n \"Part\n \"Category\"\n

\n\n> โญ **If this agent-first tool helps your workflow, please star the repo.** Stars make this tooling easier for other builders to discover and help Delx keep shipping open infrastructure.
\n> ๐Ÿงฑ Part of the [Delx agent stack](https://github.com/davidmosiah) — 15 open-source MCP servers across **body, reach and coordination**.\n\n---\n\n## HTTP (v2 stateless)\n\nDefault is **stdio**. Optional Streamable HTTP โ€” no session id, JSON responses, loopback only:\n\n```bash\nnpx -y -p short-video-agent-kit short-video-mcp --http\n# GET http://127.0.0.1:3033/health\n# POST http://127.0.0.1:3033/mcp (sessionless)\n```\n\nEnv: `SHORT_VIDEO_MCP_HOST`, `SHORT_VIDEO_MCP_PORT`, `SHORT_VIDEO_MCP_TRANSPORT=http`.\n\n\n\n\nProvider-neutral short-form AI video toolkit for agents. It gives Codex, Claude, Cursor, Hermes, OpenClaw and other MCP clients one interface for building dry-run payloads and, when explicitly enabled, generating vertical video through Sora/OpenAI, Gemini Veo, xAI/Grok and Seedance/PiAPI-style providers.\n\nUse it when an agent needs one safe interface for prompt-to-video payload validation and optional paid generation across multiple providers.\n\n## Why It Is Agent-First\n\nVideo generation can be expensive and prompt-sensitive. This package makes agents start with safe steps:\n\n- inspect provider readiness\n- return privacy boundaries\n- build payloads without spending credits\n- require `--live` or `SHORT_VIDEO_DRY_RUN=false` before provider calls\n- keep prompts and outputs in local user-controlled paths\n\n## Install\n\n```bash\nnpm install -g short-video-agent-kit\n```\n\nOr run directly:\n\n```bash\nnpm exec --yes --package=short-video-agent-kit -- short-video-agent-kit doctor\n```\n\n## Quickstart\n\nNo API key needed to try it โ€” generation is dry-run by default, so the kit returns the exact provider-neutral plan it *would* send without spending a credit.\n\nBuild a Sora plan for an 8-second vertical teaser:\n\n```bash\nshort-video-agent-kit generate \\\n --provider openai_sora \\\n --prompt \"Vertical 8-second product teaser for a minimalist water bottle, soft studio light, slow dolly-in\" \\\n --output ./output/teaser.mp4\n```\n\nReal output (no provider call, no credits spent):\n\n```json\n{\n \"ok\": true,\n \"dry_run\": true,\n \"next_step\": \"Pass --live or set SHORT_VIDEO_DRY_RUN=false to call the provider API.\",\n \"provider\": \"openai_sora\",\n \"endpoint\": \"POST /v1/videos\",\n \"payload\": {\n \"model\": \"sora-2\",\n \"prompt\": \"Vertical 8-second product teaser for a minimalist water bottle, soft studio light, slow dolly-in\",\n \"seconds\": \"8\",\n \"size\": \"720x1280\"\n }\n}\n```\n\nSame prompt, different provider โ€” the plan re-targets the endpoint and parameter shape for you. `payload` returns just the plan (no `dry_run` wrapper):\n\n```bash\nshort-video-agent-kit payload --provider gemini_veo --prompt \"Same teaser, 9:16, cinematic\"\n```\n\n```json\n{\n \"provider\": \"gemini_veo\",\n \"endpoint\": \"POST /models/{model}:predictLongRunning\",\n \"payload\": {\n \"instances\": [\n {\n \"prompt\": \"Same teaser, 9:16, cinematic\"\n }\n ],\n \"parameters\": {\n \"aspectRatio\": \"9:16\",\n \"durationSeconds\": 8\n }\n }\n}\n```\n\nCheck which providers are wired up (keys are detected, never printed):\n\n```bash\nshort-video-agent-kit doctor\n```\n\n```json\n{\n \"ok\": false,\n \"dry_run\": true,\n \"providers\": {\n \"openai_sora\": { \"configured\": false, \"env_keys\": [\"OPENAI_API_KEY\"], \"models\": [\"sora-2\"] },\n \"gemini_veo\": { \"configured\": false, \"env_keys\": [\"GEMINI_API_KEY\", \"GOOGLE_API_KEY\"], \"models\": [\"veo-3.1-fast-generate-preview\"] },\n \"xai_grok\": { \"configured\": false, \"env_keys\": [\"XAI_API_KEY\"], \"models\": [\"grok-imagine-video\", \"grok-imagine-image\"] },\n \"seedance_piapi\": { \"configured\": false, \"env_keys\": [\"PIAPI_KEY\", \"SEEDANCE_API_KEY\"], \"models\": [\"seedance-2-fast-preview\"] }\n },\n \"output_dir\": \"./output\",\n \"next_steps\": [\n \"Set one provider key: OPENAI_API_KEY, GEMINI_API_KEY, XAI_API_KEY or PIAPI_KEY.\"\n ]\n}\n```\n\nWhen you are ready to actually render, set a provider key and re-run `generate` with `--live` (or `SHORT_VIDEO_DRY_RUN=false`).\n\n## CLI\n\n```bash\nshort-video-agent-kit manifest --client codex\nshort-video-agent-kit doctor\nshort-video-agent-kit privacy-audit\nshort-video-agent-kit payload --provider gemini_veo --prompt-file prompt.txt\nshort-video-agent-kit generate --provider openai_sora --prompt \"Vertical product teaser\" --output ./output/teaser.mp4\nshort-video-agent-kit generate --provider openai_sora --prompt-file prompt.txt --output ./output/teaser.mp4 --live\n```\n\nSupported providers:\n\n- `openai_sora`\n- `gemini_veo`\n- `xai_grok`\n- `seedance_piapi`\n\n## MCP\n\n```bash\nshort-video-mcp\n```\n\nHTTP transport:\n\n```bash\nSHORT_VIDEO_MCP_TRANSPORT=http short-video-mcp\n```\n\nHermes-style config:\n\n```yaml\nmcp_servers:\n short_video:\n command: npx\n args: [\"-y\", \"short-video-agent-kit\"]\n sampling:\n enabled: false\n```\n\nRecommended first calls:\n\n1. `short_video_connection_status`\n2. `short_video_privacy_audit`\n3. `short_video_build_payload`\n4. `short_video_generate`\n\n## Agent Surfaces\n\n| Tool | Purpose |\n|---|---|\n| `short_video_agent_manifest` | Install/runtime guidance for Codex, Claude, Cursor, Hermes and OpenClaw |\n| `short_video_connection_status` | Provider readiness without API keys |\n| `short_video_privacy_audit` | Prompt, output and reference-asset boundaries |\n| `short_video_build_payload` | Provider-specific payload without paid generation |\n| `short_video_generate` | Dry-run by default, live only when explicitly requested |\n\n## Copy-Paste Agent Prompt\n\n```text\nUse short-video-agent-kit. First call short_video_connection_status and short_video_privacy_audit.\nBuild the payload before generation. Only set live=true if I explicitly confirm a paid provider call.\n```\n\n## Configuration\n\nCopy `.env.example` to `.env` and fill only the provider keys you plan to use. `.env`, `output/` and `.agent-data/` are ignored by Git.\n\n## Safety Model\n\n- Dry-run is the default.\n- API keys are never returned by tools.\n- Paid generation requires `--live`, MCP `live=true`, or `SHORT_VIDEO_DRY_RUN=false`.\n- Reference images must be user-owned or licensed.\n- Outputs are written to local paths controlled by the user.\n\n## Development\n\n```bash\nnpm install\nnpm test\nnpm run check\n```\n\n---\n\n## ๐Ÿ“ง Contact & Support\n\n- ๐Ÿ“จ **support@delx.ai** โ€” general questions, integration help, partnerships\n- ๐Ÿ› **Bug reports / feature requests** โ€” [GitHub Issues](https://github.com/davidmosiah/short-video-agent-kit/issues)\n- ๐Ÿฆ **Updates** โ€” [@delx369](https://x.com/delx369) on X\n- ๐ŸŒ **Site** โ€” [wellness.delx.ai](https://wellness.delx.ai)\n\n", "bytes": 8633, "sha": "6dac5dba6f0cfd40c557cce82a12718630d5edfd07cb5f013f4eff4cdff5a9e9", "repo_slug": "davidmosiah/short-video-agent-kit", "fonte": "repo", "truncated": false, "api": "https://agentalog.com/api/listings/mcp_io_github_davidmosiah_short_video_agent__18ff44bb/readme" }