agentleFS
postoncehqpostoncehq/postonce-skills/skills/post-everywhere/SKILL.md

post-everywhere

Skill1 starsChanged 31 days agoMITView on GitHub

Turn one idea, draft, blog post, link or video into a native post for every connected platform (LinkedIn, X, Threads, Bluesky, Facebook, Instagram, TikTok, YouTube, Pinterest) and publish or schedule them all in one PostOnce request. Use when the user asks to crosspost, repurpose, "post this everywhere", share something on all their socials, or adapt one post for several platforms.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Call list_active_accounts (or list_accounts) and read each account's capabilities and media_requirements.

    line 15
  2. Distinctive:

    Read the source (their text, a URL they gave you, a transcript).

    line 19

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Post everywhere

One source, one request, a native post on each platform. PostOnce publishes to every selected account from a single create_post call; each target gets its own copy through content_override and its own settings through platform_options. This skill decides what each platform should get, then sends it once.

1. Pick the destinations

Call list_active_accounts (or list_accounts) and read each account's capabilities and media_requirements. Use the accounts the user named; if they said "everywhere", propose every active account and let them remove any. Never guess IDs.

2. Find the one idea

Read the source (their text, a URL they gave you, a transcript). Pick the single most useful or surprising point; one post carries one idea. Keep facts, numbers and quotes exactly as the source has them. Don't invent results.

3. Write a native version per platform

Same idea, different shape. Start each version with its own first line; never paste one caption everywhere.

PlatformLimitWhat works
LinkedIn3,000First 2 lines decide the "see more" tap. Short lines, a stance, one question at the end. Link in the body is fine; many prefer the first comment.
X280 (4,000 with Premium)One sharp line. Numbers and specifics. Up to 4 images, 1 GIF or 1 video.
Threads500Conversational, a little looser than X. Up to 10 images or videos in a carousel.
Bluesky300Plain and direct. Links, #tags and @mentions become rich text.
Facebook Page63,206Warm, first person plural, a clear call to action. Up to 10 photos; video posts as a Reel.
Instagram2,200Needs media. Hook in line 1, value, then 3 to 5 specific hashtags. Carousel of 2 to 10, or one video as a Reel.
TikTok2,200Needs media: one video, or a photo slideshow. Search-style caption with the topic words people type; up to 5 hashtags.
YouTubetitle 100, description 5,000Needs a video. Put the title in platform_options.title; a vertical video under 3 minutes becomes a Short.
Pinteresttitle 100, description 500Needs an image or video. Keyword-led title and description, link to the source page.

The server doesn't reject copy that's too long: it cuts it off. Count characters for each version before sending.

4. Match media to each destination

List what media the user has. For each target, check its media requirements. Platforms that need media and have none: drop them from this post and say so, or ask for an image or video. Don't mix images and video in one post where the platform takes one type; extras are dropped silently. For a local file, upload it with create_upload_url and get_media (see the postonce skill); the hosted server can't read file paths.

5. Confirm, then send once

Show the user a short table: platform, account, the exact text, media, time. Ask for one confirmation covering all of it. Then call create_post once:

{
  "content": "<shared fallback text>",
  "publish_at": "2026-10-06T09:00:00-04:00",
  "media": [{"type": "image", "url": "<public URL>"}],
  "targets": [
    {"account_id": "<linkedin id>", "content_override": "<LinkedIn version>"},
    {"account_id": "<x id>", "content_override": "<X version>"},
    {"account_id": "<youtube id>", "content_override": "<description>", "platform_options": {"title": "<title>"}}
  ]
}

Omit publish_at to publish now. Staggered times need separate posts, one per time.

6. Report what actually happened

Call get_post with the returned ID and report each target on its own line: published with URL, scheduled, processing, or failed with the error. A created post isn't a delivered post. Retry only a failed target, using the recovery steps in the postonce skill; never re-send the whole post.

For deeper platform writing (carousels, scripts, threads, thumbnails), the per-platform plugins in postoncehq/plugins add skills for each network. For every future post to crosspost automatically, set up a workflow with the postonce skill instead.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.