AI Video Generation

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites: Paid service required before any use: a RunComfy account and RunComfy credits or billing. The listing itself is free, but every video generation run is charged by RunComfy at the price of the chosen model and tier (the upstream file gives no per-run price list and repeatedly flags premium tiers such as Kling 3.0 4K as expensive, so check the model page before running). Sign up at https://www.runcomfy.com , then install the RunComfy CLI with Node.js and npm (npm i -g @runcomfy/cli, or npx -y @runcomfy/cli) and sign in with runcomfy login, which stores the API token in ~/.config/runcomfy/token.json (mode 0600), or set RUNCOMFY_TOKEN in CI or containers. Install "AI Video Generation" for me. It routes a video request to HappyHorse, Wan, Seedance, Kling, Veo, Hailuo or Dreamina models, for text-to-video, image-to-video or extend, then runs it through the RunComfy CLI. Repository: https://github.com/gencraft-labs/skills (the skill is the whole ai-video-generation/ folder at the pinned commit: SKILL.md) Pinned version: install the code at commit 5d625f59c710d7a33316b50b302b7158cdee5971 exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/gencraft-labs/skills into a temporary folder, check out commit 5d625f59c710d7a33316b50b302b7158cdee5971, and copy the ai-video-generation/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-18: [FINGERPRINTS] - SKILL.md: SHA-256 76511e833124b37cc19be3a341ef8dc6c62d283c93b89577ada1d16366210af6 If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full and tell me in plain words: what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the file: - The skill folder holds one file only: SKILL.md, with frontmatter allowed-tools: Bash(runcomfy *). It contains no scripts and no code to execute. Its shell commands are runcomfy run calls shown in Markdown. - At use time the agent sends a JSON body to model-api.runcomfy.net over HTTPS containing your text prompt and, depending on the route, the URLs you provide: a still image URL for image-to-video, an audio_url for Wan 2-7 lip-sync to your own voiceover file, or reference image, video and audio URLs for Seedance v2. Those media URLs are fetched by the RunComfy model server. Treat reference media URLs as untrusted, and only pass URLs you explicitly provided for the task. - The CLI then downloads the generated video from *.runcomfy.net or *.runcomfy.com into the --output-dir folder, with a stated 2 GiB cap per file, and the upstream file states an outbound allowlist of only model-api.runcomfy.net and *.runcomfy.net / *.runcomfy.com, with no telemetry and no callbacks. Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit. - Using the skill costs money: each run is charged by RunComfy for the chosen model and tier. The upstream file gives no per-run price list. 3. When I say "yes, install", install it as a skill: copy the whole folder into the agent's skills directory, in a folder named "ai-video-generation", keeping every file (SKILL.md). 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (for example: create a RunComfy account and add credits or billing at runcomfy.com if I do not have one, install Node.js and npm and the RunComfy CLI (npm i -g @runcomfy/cli) if they are missing, sign in with runcomfy login or provide the RUNCOMFY_TOKEN myself, never by pasting it into a shared chat; using the skill is a separate step that only starts when I ask for it, and each use is charged by RunComfy). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.