Blog post
Transform a video into a rich blog post (HTML + embedded markdown) with sections, thumbnails, and key takeaways. Use when the user wants to turn a video, talk, demo, or tutorial into a written article or blog content. Takes one source: a local video file, URL, or cloudglue:// file URI (e.g. cloudglue://files/<id>). Runs the built-in tinycloud "blog-post" workflow; requires the tinycloud CLI configured with a Cloudglue API key (analysis runs through the user's Cloudglue account).From its SKILL.md
npx -y skills add cloudglue/tinycloud --skill blog-postAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 1 stars1 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.
SKILL.md
2.7 KB, as published. Nobody here has run it
Video → blog post
This skill is a thin wrapper around the blog-post workflow recipe bundled
inside the tinycloud binary (watch → extract → render).
Run
-
Check the CLI. If the general
tinycloudskill is installed alongside this one, run itsscripts/preflight.sh. Otherwise verify directly:tinycloud setup --check --json # ready when data.ok == trueMissing CLI:
npm install -g @cloudglue/tinycloud(see https://tinycloud.sh). Missing key:tinycloud setup cloudglue --api-key <key>. -
Confirm the recipe is available (free, no cloud calls):
tinycloud workflow validate blog-post --json -
Run it with the user's source. The analysis steps run through the configured Cloudglue API key — if the user has not clearly asked to run it, show the step plan first with
tinycloud workflow plan blog-post $source --json(free).tinycloud workflow blog-post $source --jsonUseful params:
--param segment=chapters(default; semantic section anchors) or--param segment=uniform:20for a lighter pass;--param out=<path>to control the HTML location.
Read the result
Parse the single JSON envelope from stdout (machine output; logs are stderr):
- Success:
status == "ready"anddata.status == "completed". - The article path is
data.outputs.html(also indata.artifacts[]). Default:./tinycloud-output/runs/<data.run_id>/blog-post.html. - Report the HTML path; offer
tinycloud publish <html> --name blog-post --visibility private --jsonto host it as a shareable page. Share the returneddata.url(fresh content can take ~1 min to appear there;data.version_urlis live immediately, so a brief 403 atdata.urlis not a failure).
Any other status (needs_credentials, needs_upload, pending, paused,
error) or data.status of partial/failed: stop, report the envelope's
error.message, and follow its setup / resume / next hints. The general
tinycloud skill (if installed) documents full status handling.