agentleFS
fajrisilmi12-cyberfajrisilmi12-cyber/hermes-socmed-function/skills/zernio-social-posting/SKILL.md

zernio-social-posting

Skill20 starsChanged 3 months agoMITView on GitHub

Post to 16 social platforms via Zernio + Cloudinary CDN.

  • Reads credentials

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    Do NOT generate images from a prompt for this user unless they explicitly ask.

    line 30
  2. Very specific:

    Only supports post content type — no story/reel.

    line 141
  3. Very specific:

    Never use commas in Cloudinary transform URLs: when posting via Zernio. Always slash-chain.

    line 165

≈ “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

Zernio + Cloudinary Social Media Posting

Post to 16 platforms via a single API: Zernio handles publishing, Cloudinary hosts + optimizes the media. Free tier covers 1-2 accounts on Zernio (no credit card) and 25 credits/month on Cloudinary.

When to use this skill

  • User wants to post an image/video to Instagram, Facebook, TikTok, X, etc.
  • User has a media URL and wants to publish to a social platform
  • User wants to schedule a post or cross-post to multiple platforms
  • User wants Story (24h ephemeral) or Reel (vertical video) support
  • User already has a Zernio account and an Instagram Business/Creator account

User preference: never auto-generate images

This user (Fajri) always supplies the source image themselves — usually via WhatsApp or Discord attachment. The pipeline is:

  1. User sends image + caption
  2. Save to /tmp/ (or wherever the platform's bridge drops it)
  3. Run post_to_ig.py /tmp/<file> "<caption>"
  4. Reply with the live IG URL

Do NOT generate images from a prompt for this user unless they explicitly ask. Pair with the image-gen-9router skill only when they request text-to-image. The default request "post this to IG" means "I have an image, please upload and post it" — not "make me an image and post it."

The full text-to-image-then-post flow (e.g. "research Linux, generate an infographic, post it") is a separate, opt-in workflow that combines this skill with image-gen-9router.

What this skill assumes

  • Zernio account at https://zernio.com/signup (no CC for first 1-2 accounts)
  • At least one social account connected via Zernio OAuth
  • Instagram account must be Business or Creator (Personal accounts cannot use Graph API)
  • Cloudinary account for media CDN (free tier: 25 credits/month)
  • Env vars in ~/.hermes/.env:
    • ZERNIO_API_KEY=sk_...
    • CLOUDINARY_CLOUD_NAME=...
    • CLOUDINARY_API_KEY=...
    • CLOUDINARY_API_SECRET=...
    • Optional: ZERNIO_IG_ACCOUNT_ID=... (default for single-account flow)

The pipeline

Local file or URL
  ↓
Cloudinary upload (auto resource_type=image|video)
  ↓
Build URL with slash-chained transforms (NEVER commas — see pitfall below)
  ↓
POST to Zernio /v1/posts with bearer auth
  ↓
Zernio publishes to social platform(s)
  ↓
Live URL returned

Slash-chained Cloudinary URLs (the working pattern)

Cloudinary transform syntax has two flavors. Zernio's URL parser only accepts the slash-chained form. See references/cloudinary-comma-bug.md for the full transcript of the failure.

Broken (commas mis-parsed as multi-URL separator):

https://res.cloudinary.com/X/image/upload/w_1080,h_1080,c_fill,q_auto,f_auto/v123/id.jpg
              ↑ Zernio reads "h_1080" as second URL → 400 Invalid media URL

Works (slash-chained, equivalent transform):

https://res.cloudinary.com/X/image/upload/w_1080/h_1080/c_fill/q_auto/f_auto/v123/id.jpg

For video:

https://res.cloudinary.com/X/video/upload/q_auto/v123/id.mp4

Zernio content types (Instagram)

TypeZernio fieldNotes
Feed postomit contentTypedefault; image, video, carousel
StorycontentType: "story"24h ephemeral; captions not shown
ReelcontentType: "reel"vertical 9:16 video, ≤90s; auto-applied to single video without the field
{
  "content": "caption with #hashtags",
  "mediaItems": [{"type": "image", "url": "<cloudinary slash-chained URL>"}],
  "platforms": [{
    "platform": "instagram",
    "accountId": "<ZERNIO_IG_ACCOUNT_ID>",
    "platformSpecificData": {"contentType": "story"}
  }],
  "publishNow": true
}

Available templates

  • templates/post_to_ig.py — full one-shot post: upload to Cloudinary, post via Zernio, wait for live URL, print result. Supports feed/story/reel. Auto-loads ~/.hermes/.env.
  • templates/cloudinary_helper.py — Cloudinary SDK wrapper: upload + slash-chained IG URL + usage stats. Auto-loads ~/.hermes/.env.

Usage

python3 templates/post_to_ig.py photo.jpg "caption #tags"
python3 templates/post_to_ig.py photo.jpg "caption" --content-type story
python3 templates/post_to_ig.py video.mp4 "caption" --content-type reel

# Schedule for future publish (>= a few hours ahead)
python3 templates/post_to_ig.py photo.jpg "caption" --platform threads \
  --schedule "2026-08-30T09:00:00" --tz "Asia/Jakarta"

Multi-platform support (IG + Threads + more)

The script defaults to Instagram. Use --platform to target other platforms Zernio supports (threads, facebook, x, tiktok, linkedin, etc.). Each platform has its own ZERNIO_*_ACCOUNT_ID env var.

# Post to Threads instead of Instagram
python3 templates/post_to_ig.py photo.jpg "Thread about X" --platform threads

Per-platform caveats:

  • Threads: 500 char caption limit (hard cap from Meta API). Only supports post content type — no story/reel.
  • Instagram: full feed/story/reel support. 2200 char caption.

Scheduling

--schedule accepts an ISO 8601 datetime plus --tz for the timezone (default Asia/Jakarta). Zernio queues the post and publishes automatically. Threshold: schedule at least 1 hour ahead; Zernio treats shorter windows as "publish now".

# Schedule for tomorrow morning WIB
python3 templates/post_to_ig.py photo.jpg "Morning thought" --schedule "2026-08-30T09:00:00"

# Schedule for 3 days from now, 6pm Eastern
python3 templates/post_to_ig.py photo.jpg "Friday post" --schedule "2026-09-01T18:00:00" --tz "America/New_York"

Behind the scenes, the script sets publishNow: false and adds scheduledFor + timezone to the Zernio payload. The status will read scheduled (not published) until the publish time.

Pitfalls

  • Never use commas in Cloudinary transform URLs when posting via Zernio. Always slash-chain. See references/cloudinary-comma-bug.md.
  • IG account must be Business or Creator. Personal accounts will fail at OAuth with no helpful error. Switch in IG app: Settings → Account type and tools → Switch to professional account.
  • Zernio 401 with Invalid API key usually means the bearer token is the wrong env var. zernio auth:set --key <value> writes to ~/.config/zernio/auth.json; the Python script reads from ZERNIO_API_KEY env var (or auto-loads from ~/.hermes/.env).
  • IG token lasts 60 days; Zernio will ask for reconnect. Watch for token_expired errors.
  • n (count) field is provider-dependent. Some providers (gemini nano-banana, huggingface) ignore it.
  • Story captions are NOT shown on Instagram (Graph API limit, not Zernio).
  • Reel shareToFeed — default false. Set true in platformSpecificData to also appear in main feed.
  • First-call latency — Zernio returns status: publishing then polls to published. The Python template polls /v1/posts/{id} for up to 60s for live URL.

Reference docs

  • references/cloudinary-comma-bug.md — full transcript of the URL parser failure
  • references/env-loader-pattern.md — Python pattern for auto-loading ~/.hermes/.env

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.