agentsclimarketplace

Vibeknow create

Skill vibeknow/cli/skills/vibeknow-create

Generate videos from documents/URLs/files, track video task progress, download results, list voice templates. Use when: user wants to create a video, check task status, download video, or browse voices.From its SKILL.md

Install
npx -y skills add vibeknow/cli --skill vibeknow-create

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

3 things to look at

  • reads credentialsReads from 1 credential source: `VIBEKNOW_TOKEN`.
  • 6 stars6 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
  • runs commandsInstructs the agent to run 7 commands, including `vibeknow create --from <source>` and 6 more.

SKILL.md

5.8 KB, ~1.3k tokens by cl100k_base, as published. Nobody here has run it

vibeknow-create

TRIGGER

  • User wants to generate a video from a document, URL, or file
  • Check video task status or wait for completion
  • Download a rendered video
  • List available voice templates

SKIP

  • Document upload/status only (no video) → use vibeknow-doc
  • Auth, profile, config, diagnostics → use vibeknow-core

Core Concepts

  • Hero command: vibeknow create --from <source> resolves input → uploads if needed → submits to figlens pipeline → streams progress → returns video URL.
  • --from accepts 3 input types: doc_id (used directly), URL (auto-uploaded to vectoria), local file path (auto-uploaded).
  • Sync vs async: Default is sync (blocks until done). --async returns task_id + session_id immediately.
  • NDJSON event stream: --output ndjson emits structured progress events (schema_version: "1"). See events.md.
  • 6 pipeline stages: parse → outline → storyboard → tts → render → publish.
  • session_id: All video subcommands require both <task_id> and --session-id. These are returned by create.

Quick Reference

CommandDescription
vibeknow create --from <source>Generate a video (sync by default)
vibeknow video status <task_id> --session-id <sid>Get task status
vibeknow video wait <task_id> --session-id <sid>Stream progress, block until done
vibeknow video download <task_id> --session-id <sid>Download rendered video
vibeknow voice listList available voice templates

For full flags and output examples, see commands.md.

Common Tasks

Generate a video (sync, simplest path)

vibeknow create --from slides.pdf
# Blocks until done, prints video URL

Generate with specific voice

vibeknow voice list                              # find voice ID
vibeknow create --from slides.pdf --voice v_warm_female

Async submit, then follow up

# Submit and exit immediately
vibeknow create --from https://example.com/doc --async
# Output: task_id=t_xxx session_id=s_yyy

# Later: check status
vibeknow video status t_xxx --session-id s_yyy

# Or: wait for completion
vibeknow video wait t_xxx --session-id s_yyy

Agent mode (NDJSON streaming)

vibeknow create --from doc_abc --output ndjson
# Each line is a JSON event: task.submitted, stage.started, stage.progress, ...
# Terminal event: task.succeeded (with video_url) or task.failed

Download the result

vibeknow video download t_xxx --session-id s_yyy
# Default output: <session_id>.mp4

vibeknow video download t_xxx --session-id s_yyy --output ./my-video.mp4
vibeknow video download t_xxx --session-id s_yyy --output ./my-video.mp4 --overwrite

Exit Code Handling

ExitMeaningAgent Action
0SuccessExtract video_url from output
1General errorRead stderr
2Invalid argumentsFix command syntax
3Auth errorRun vibeknow auth status to inspect credential source. Re-login with vibeknow auth login (interactive) or set VIBEKNOW_TOKEN. See vibeknow-core for profile/diagnostics if installed.
4Task failed, retryableRe-submit the same create command
5Task failed, not retryableReport error to user, do not retry
6Stream interrupted, task status unknownvibeknow video wait <task_id> --session-id <sid> to reconnect. Do NOT re-submit.
130User interrupt (SIGINT)—

For detailed error handling and recovery, see errors.md and recipes.md.

NDJSON Event Summary

Events share common fields: schema_version, ts, type.

Key events (pipeline engine):

EventExtra FieldsMeaning
node.startedstage, node, messagePipeline node begins
node.succeededstage, node, messageNode done
node.failedstage, node, messageNode failed (not necessarily terminal — wait for task.failed)
task.succeededsession_id, video_url, duration_msTerminal: video ready
task.failedcode, message, retryableTerminal: task failed (retryable=true → exit 4, false → exit 5)

Agent engine (--engine agent) replaces node.started/succeeded/failed with node.progress carrying status + message, and omits duration_ms from task.succeeded.

See events.md for the complete field reference, engine differences, and parsing examples.

References

  • commands.md — Full flag reference for all commands
  • events.md — NDJSON task event schema
  • errors.md — Exit codes, error codes, Error Object schema
  • recipes.md — Advanced: retry, recovery, batch, NDJSON parsing

What ships with it: 4 files

15.2 KB alongside SKILL.md

references/

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.