Krillinai Pipeline
Apps & Automationkrillinai-pipelineKrillinai Pipeline is an Agent Skill for apps & automation published in the krillinai/KrillinAI repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Use when planning multi-stage KrillinAI CLI workflows such as subtitle plus TTS plus horizontal or vertical render; current pipeline execution is a reserved/planned surface unless implementation is wired in.
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Inspect the Agent Skill at https://github.com/krillinai/KrillinAI/blob/HEAD/skills/krillinai-pipeline/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/krillinai-pipeline, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
KrillinAI Pipeline
Use this skill for end-to-end or multi-stage planning. In the current CLI, pipeline has output planning support but execution may not be wired in; prefer individual commands for real work unless verified.
Command
./build/krillinai-cli pipeline --outputs "subtitle,tts,vertical-bilingual"
Outputs Values
| Output | Stage |
|---|---|
subtitle | Generate subtitle files |
tts | Generate dubbing |
horizontal-bilingual | Render landscape bilingual video |
horizontal-dubbed | Render landscape dubbed video |
vertical-bilingual | Render portrait bilingual video |
vertical-dubbed | Render portrait dubbed video |
cover | Generate cover |
When To Use Pipeline
Use pipeline when the user asks for a complete result and the CLI implementation supports executing it. Prefer individual commands now for real runs, debugging, testing, or recovering a failed stage.
Recovery Strategy
- If
subtitlesucceeds and render fails, rerun only render with the same--workdir. - If TTS fails, keep subtitle outputs and rerun
ttsafter fixing provider config. - Read
krillinai_manifest.jsonbefore rerunning to avoid repeating expensive work.
Verification
- If the command returns
unsupported_command, run the stages individually. - Confirm stdout JSON success when pipeline execution is wired in.
- Confirm manifest stages for requested outputs are marked successful.
- Inspect final requested media/cover files, not only intermediate outputs.
- For shared CLI details, read
skills/krillinai-cli/references/cli-contract.md.