agentleFS
PerufitlifePerufitlife/postwire-skills/skills/postwire/SKILL.md

postwire

Skill0 starsChanged 30 days agoMIT-0View on GitHub

Publish and schedule to TikTok, Instagram, YouTube, LinkedIn, Bluesky, Mastodon, Facebook Pages, Telegram and Discord through one PostWire API call — writing a separate, native version of the post for each network instead of sending identical text everywhere. Use when the user wants to publish or schedule one idea across several platforms, wants per-network captions, or wants to know why a post was rejected before it is sent.

  • Reads credentials
  • Sends data out

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before either call, show the user:

    line 85

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

PostWire

Publish one idea to many social networks, with a different post written for each one.

Accounts are connected once through OAuth in the PostWire dashboard, so there is no per-platform developer app, review queue, or token refresh to maintain. One API key covers every network.

When to Use This Skill

  • The user wants the same idea published to several networks, but written properly for each
  • The user wants a post scheduled for later, or a week of posts queued from one topic
  • The user asks why a post failed, or whether a post will fail before sending it
  • The user is publishing video to TikTok or YouTube from an automation

What Makes This Different

Most multi-posting APIs take one string and send it to every platform. That reads badly and gets suppressed: a LinkedIn post written like a tweet performs like neither. PostWire has a /api/generate step that writes a caption per network — length, tone, hashtags and link handling — from a single prompt, which you then pass to publish or schedule.

It also refuses posts that are going to fail, before they are queued rather than after the platform rejects them:

networkrequires
TikTok, YouTubea video (video_url)
Instagrama photo or a video
everything elsetext is enough

Prerequisites

  • An account at postwire.io — free tier: 1 brand, 20 posts/month, 2 networks per post, no card
  • Networks connected in the dashboard (one OAuth click each)
  • POSTWIRE_API_KEY in the environment

Auth is Authorization: Bearer $POSTWIRE_API_KEY.

One gotcha worth knowing: the dashboard shows a preview key immediately after signup, before the email is confirmed. That key cannot publish. If the API answers "This is a preview key" (code: "preview_key"), confirm the email and take the key from API & MCP in the dashboard.

How to Use

Base URL: https://postwire.io/api

Check what is connected

curl https://postwire.io/api/me -H "Authorization: Bearer $POSTWIRE_API_KEY"

Returns the plan, the monthly usage, and connections[] — publish only to networks listed there, otherwise the call fails with code: "not_connected" and the offending platforms.

Write a native caption per network

curl -X POST https://postwire.io/api/generate \
  -H "Authorization: Bearer $POSTWIRE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "We shipped streaming uploads: 1GB videos now go through in 11 seconds",
    "platforms": ["linkedin", "instagram", "bluesky"]
  }'

Returns { "drafts": { "linkedin": {...}, "instagram": {...}, "bluesky": {...} } }. That object is what per_platform takes below — but show it to the user first, for the reason in the next section.

Confirm Before Publishing

/api/post and /api/schedule are visible writes: they put text under the user's own name, in public, where other people see it. Treat them the way you would treat sending an email as them.

Before either call, show the user:

  • the exact text that will go out, per network — not the prompt that generated it
  • which networks it is going to
  • when it publishes (now, or the exact run_at)

and wait for a yes.

This matters most right after /api/generate. Those drafts are model output: passing them straight into per_platform without showing them means the user finds out what they said by reading their own timeline. Generate, show, confirm, then post.

Publish now

curl -X POST https://postwire.io/api/post \
  -H "Authorization: Bearer $POSTWIRE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "platforms": ["linkedin", "bluesky"],
    "per_platform": { "linkedin": {"text": "..."}, "bluesky": {"text": "..."} }
  }'

Without per_platform, a single text is used everywhere — simpler, but it is the behaviour this tool exists to avoid.

The response is per platform, and one failure does not cancel the rest:

{ "results": [ { "ok": true, "platform": "linkedin", "id": "..." },
               { "ok": false, "platform": "bluesky", "error": "..." } ] }

Schedule instead

curl -X POST https://postwire.io/api/schedule \
  -H "Authorization: Bearer $POSTWIRE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "run_at": "2026-09-01T09:00:00Z",
    "platforms": ["linkedin", "instagram"],
    "photo_url": "https://example.com/image.jpg",
    "label": "Launch announcement"
  }'

run_at is ISO 8601. The schedule endpoint validates connections and the media rules up front, so a post that cannot work is rejected now instead of failing quietly at 7am.

Fill a week from one topic

curl -X POST https://postwire.io/api/week \
  -H "Authorization: Bearer $POSTWIRE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "topic": "what we learned running a one-person SaaS", "platforms": ["linkedin", "bluesky"], "days": 5 }'

Five angles on one topic, queued across five days. Text-only networks only — the video platforms would reject every one of them, and the endpoint says so rather than queueing them to fail.

Errors Worth Handling

codemeaningfix
preview_keykey from before email confirmationconfirm email, use the key in API & MCP
email_unverifiedmailbox not confirmed yetopen the emailed link
not_connectedplatform not connected; platforms[] lists whichconnect it in the dashboard
media_requiredTikTok/YouTube without video, Instagram without mediaadd video_url or photo_url
429monthly plan limit or a burst guardcheck usage in /api/me

Media Notes

  • Media is passed by URL (video_url, photo_url); PostWire fetches and forwards the bytes.
  • TikTok and YouTube videos are streamed in chunks, so they can be up to 1GB. Other networks and hosts that do not send Content-Length are capped at 80MB, because the file has to be held in memory.
  • POST /api/media/upload-url returns a signed URL to upload a file directly (up to 1 GB; files over 50 MB send "multipart": true and PUT each returned part) if you do not have somewhere to host it.

Also Available

PostWire ships an MCP server, so an agent can call these as tools rather than as HTTP requests, and there are n8n templates for the same flows. Both are linked from the dashboard.

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.