Skip to content

KrillinAI Pipeline

Validate a KrillinAI video output plan and guide running its individual stage commands

Stars
14,089
Licence
Apache-2.0
Files
1 file

What it does

This skill teaches an AI agent how to plan video processing jobs for the KrillinAI command-line tool, which handles subtitles, dubbing, rendering landscape or portrait videos, and cover images. The agent can check a requested set of outputs with a dry-run, which validates the names without doing any real work. For actual processing, the skill tells the agent to run each stage command separately and to use the manifest file and output checks to recover from failures and confirm results.

Great for

  • Creators who produce subtitled, dubbed or bilingual versions of videos with KrillinAI and want an agent to plan the steps
  • Developers scripting KrillinAI batch jobs who want a dry-run check before running stages
  • Teams that need a checklist for verifying that requested media and cover files were actually produced

Skip it if

  • Anyone expecting one command to run the whole pipeline end to end: the skill states that non-dry-run pipeline execution returns unsupported_command
  • Users without KrillinAI installed, since the skill only orchestrates an existing CLI and does not provide it
  • Sensitive content workflows where media is sent to third-party speech or translation providers without review, because the skill does not describe what data leaves the machine

Try asking

  • “Check whether a plan with subtitle, tts and vertical-bilingual outputs is valid before I start processing my video”
  • “Run the subtitle stage, and if rendering fails, rerun only the render step using the same work directory”
  • “Explain which KrillinAI stages map to each output type such as horizontal-dubbed or cover”
  • “Verify that the final portrait video and cover image exist after the stages finish”
Step-by-step guide: how to use KrillinAI PipelineInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    krillinai-pipeline.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip krillinai-pipeline.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip krillinai-pipeline.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Run the dry-run first to confirm output names are valid before any expensive processing
  • Read krillinai_manifest.json before rerunning a failed stage to avoid repeating costly work
  • Check the final media and cover files directly rather than trusting the stage JSON alone

What's inside

  • SKILL.md

Redistributed unmodified from krillinai/OpenCreator under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.