Social Post logo

Social Post

Organization
publora
social-post

Use when the user wants to post or schedule to YouTube, a Facebook Page or Mastodon through Publora. YouTube needs a video on every post, Facebook publishes to Pages only, Mastodon caps at 500 characters. Not for the platforms that have their own skill.

Overview

Publisherpublora
Repositoryskills
Skill namesocial-post
Stars
62
Forks
6
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by publora on GitHub. Read the source before you install it.

Installation

Install the Social Post AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/publora/skills/tree/main/skills/social-post
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/publora/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/skills/social-post .claude/skills/social-post
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Social Post in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Social Post on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Social Post is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Social Post (YouTube, Facebook, Mastodon)

Create and schedule posts to YouTube, Facebook Pages, and Mastodon using the Publora MCP server. This skill covers platforms that work through standard posting without platform-specific complexity.

Prerequisites

Plans: Works on the free Starter plan. Current limits and pricing: publora.com/pricing.

Getting Started

  1. Create account at publora.com/register (free)
  2. Connect platforms via OAuth in Publora Dashboard:
    • YouTube: Google OAuth (requires YouTube channel)
    • Facebook: Facebook OAuth (Pages only, not personal profiles)
    • Mastodon: OAuth (mastodon.social instance)
  3. Get an API key at app.publora.com/dashboard/api, only if your agent cannot sign in with OAuth or you will use the REST API fallback below
  4. Connect your agent to the MCP server at https://mcp.publora.com/mcp:
    • Upgrading a manual connection without /mcp: Claude treats https://mcp.publora.com as a different server from the bundled one. Change that existing connection to https://mcp.publora.com/mcp in its current scope, preserving its authentication and chosen account; sign in again if requested. Alternatively, verify the bundled connection uses the intended account, then remove only the old manual entry. Do not add a second entry or delete credentials automatically.
    • Publora tools already available (the publora-skills plugin or the Publora connector in Claude): nothing to set up. Claude asks you to sign in to Publora the first time a tool runs.
    • Agents that support OAuth (Claude Code, Cursor, and the Claude and ChatGPT apps): add the server by URL and sign in, no API key needed. In Claude Code, add it and then run /mcp to sign in:
bash
claude mcp add --transport http --scope user publora https://mcp.publora.com/mcp
  • Agents without OAuth: send the API key as Authorization: Bearer sk_YOUR_API_KEY:
bash
claude mcp add --transport http --scope user publora https://mcp.publora.com/mcp \
  --header "Authorization: Bearer sk_YOUR_API_KEY"

Claude Desktop, Cursor, Codex, OpenClaw and the claude.ai connector each need a different config file or flow: see client setup for the exact path and snippet.

REST API Fallback

If MCP is unavailable, use the REST API directly:

Base URL: https://api.publora.com/api/v1

Authentication: x-publora-key header

bash
# Get connected platforms
curl -X GET "https://api.publora.com/api/v1/platform-connections" \
  -H "x-publora-key: sk_your_api_key"

# Create a post
curl -X POST "https://api.publora.com/api/v1/create-post" \
  -H "x-publora-key: sk_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "platforms": ["youtube-UCxxx", "facebook-123456", "mastodon-789"],
    "content": "Your post content",
    "scheduledTime": "2026-03-25T10:00:00Z"
  }'

Platform ID Formats:

  • YouTube: youtube-{channel_id} (e.g., youtube-UCxxx)
  • Facebook: facebook-{page_id} (e.g., facebook-112233445566)
  • Mastodon: mastodon-{id} (e.g., mastodon-456)

📖 Docs: docs.publora.com


YouTube

Video-only platform - every post requires a video file.

Platform Limits

FeatureLimit
Video title100 characters
Video description5,000 characters
Video duration12 hours
Video size256 GB
Video formatsMP4, MOV, AVI, WebM

Platform Settings (via REST API)

json
{
  "platformSettings": {
    "youtube": {
      "privacy": "public",
      "title": "My Video Title"
    }
  }
}
SettingValuesDefault
privacy"public", "unlisted", "private""public"
titlestringDerived from first 70 chars of content

Note: platformSettings is accepted by the MCP create_post and update_post tools as well as over REST. The schema is strict: a mistyped platform or key is rejected with a validation error rather than silently dropped.

YouTube Examples

Upload this tutorial video to YouTube:
"How to Build a REST API in 10 Minutes - Complete guide for beginners"
Upload this as an unlisted video (for internal sharing):
"Internal demo - Q1 dashboard features"

YouTube Restrictions

  1. Video required: YouTube does not support text-only or image posts
  2. Single video only: One video per post
  3. Processing time: Videos take seconds to minutes to process after upload
  4. Token refresh: YouTube tokens expire after ~6 months, reconnect if needed

Facebook

Pages only - posts go to Facebook Pages, not personal profiles.

Platform Limits

FeatureLimit
Post text2,200 characters (frontend) / 63,206 (API)
ImagesUp to 10 per post
Image size10 MB
Video duration45 minutes
Video size2 GB

Multi-Page Support

If you manage multiple Facebook Pages, each gets its own platform ID:

  • facebook-112233445566 (Page 1)
  • facebook-778899001122 (Page 2)

You can post to multiple pages in a single request.

Facebook Examples

Post this announcement to my Facebook Page:
"We're thrilled to announce our new product launch! Visit our website to learn more."
Post this to both my Facebook Pages:
"Holiday notice: Our offices will be closed on Monday."
Share these event photos on Facebook:
"Highlights from our company retreat! Great team, great memories."

Facebook Restrictions

  1. Pages only: Cannot post to personal profiles via API
  2. WebP auto-converted: WebP images become JPEG automatically
  3. 59-day token: Tokens auto-refresh, but may need reconnection if it fails
  4. No mixed media: A post can have images OR video, not both

Mastodon

Fediverse platform - posts federate across the decentralized network.

Platform Limits

FeatureLimit
Post text500 characters
ImagesUp to 4 per post
Image size16 MB
Video size~99 MB
Image formatsJPEG, PNG, GIF, WebP
Video formatsMP4, WebM, MOV

Mastodon Examples

Post this to Mastodon:
"Hello fediverse! We just shipped a major update to our open-source project. #opensource"
Share this screenshot on Mastodon:
"New feature: dark mode is here! #ui #darkmode"

Mastodon Restrictions

  1. mastodon.social only: Currently connects only to mastodon.social instance
  2. Public visibility: All posts are public by default
  3. No auto-threading: Unlike X/Twitter and Threads, Mastodon doesn't split long content
  4. 500-char strict: Content over 500 characters will be rejected

Available Tools (All Platforms)

create_post

Create a new post on any platform.

Parameters:

  • platforms: Array of platform IDs (e.g., ["youtube-UCxxx", "facebook-123", "mastodon-456"])
  • content: Post text
  • scheduledTime: ISO 8601 UTC datetime. Optional: omit it and the post is created as a draft. Send a future time to schedule. A time five or more minutes in the past is rejected with SCHEDULED_TIME_IN_PAST, so for immediate posting use the current time plus a minute.
  • mediaUrls: up to 10 public https image or video URLs. Publora downloads them server-side and attaches them before validation, so media and scheduling happen in one call. This is the one-shot alternative to the draft then get_upload_url then complete_media flow. Ingestion is rate-limited to 60 URLs per hour.

get_upload_url

Get presigned URL for media uploads.

Parameters:

  • postGroupId: Post ID to attach media to
  • fileName: File name
  • contentType: MIME type
  • type: "image" or "video"

complete_media

Finalize a file uploaded through get_upload_url, after the presigned PUT succeeds. Optional, because scheduling also finalizes pending media, but calling it early surfaces format and probe errors before publish. Not needed for media attached with mediaUrls.

Parameters:

  • mediaId: the id returned by get_upload_url

list_connections

List your connected accounts with their platform IDs. Call this first and copy the IDs verbatim; they are never guessable.

list_posts / get_post / update_post / delete_post

Manage scheduled and draft posts. update_post also patches content and platforms on a draft or scheduled post, so fixing a typo or retargeting no longer means delete and recreate. delete_media and prune_media_reference clean up uploaded files.


Cross-Platform Posting

Post to multiple platforms at once:

Post this video to YouTube and this text to Facebook and Mastodon:
"Announcing our new feature - watch the demo!"

Publora handles platform-specific requirements automatically:

  • YouTube gets the video with title/description
  • Facebook gets the text + video
  • Mastodon gets the text (500 char limit applies)

Before cross-posting or scheduling a batch, call account_context to check your remaining post quota and the latest date you can schedule.


Troubleshooting

PlatformErrorSolution
YouTube"Video required"Attach a video file
YouTube"Account disconnected"Reconnect via dashboard
Facebook"Page not found"Verify page connection
Facebook"Token expired"Reconnect page in dashboard
Mastodon"Content too long"Shorten to under 500 chars
Mastodon"Too many images"Reduce to 4 or fewer
AllRate limit errorsWait and retry

The 512 MB figure that used to appear here belongs to the dashboard's /media/process-video multipart endpoint. API and MCP uploads are presigned and use the platform ceilings above.

Frequently asked questions

What does the Social Post AI skill do?

Use when the user wants to post or schedule to YouTube, a Facebook Page or Mastodon through Publora. YouTube needs a video on every post, Facebook publishes to Pages only, Mastodon caps at 500 characters. Not for the platforms that have their own skill.

Why use Social Post on TypingMind?

Because you install it once and use it with any model. Social Post is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Social Post in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/publora/skills/tree/main/skills/social-post. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Social Post?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Social Post?

As many as you like. As long as a model supports skills, you can use Social Post with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Social Post AI skill free?

Yes. It is published on GitHub by publora under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇