> For the complete documentation index, see [llms.txt](https://oodlz-ai-studio.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://oodlz-ai-studio.gitbook.io/docs/inbound-agent/every-channel/social-listening/per-platform-limits.md).

# Per-Platform Limits

Markdown stripped, character limits enforced per platform, malformed replies dropped — replies post cleanly or do not post at all.

> **Clean plain text, within the platform's character budget — or no reply at all.**

Raw AI output on social platforms looks broken. Markdown asterisks render as literal characters. Hashtag headings post as visible text. Hyperlinks become bracketed URL noise. Replies that exceed Instagram's 2200-character limit get truncated mid-sentence. X replies over 270 characters fail to post entirely. The Oodlz social listening pipeline strips formatting and enforces per-platform character limits deterministically before any reply is sent.

## How It Works

Every agent response is parsed and stripped of markdown formatting on the way out — headings, bold, italic, link syntax all converted to clean plain text suitable for social rendering. Per-platform character limits are then enforced as hard ceilings before the reply is POSTed back to Zernio.

| Platform  | Character limit |
| --------- | --------------- |
| Instagram | 2200            |
| Facebook  | 2200            |
| LinkedIn  | 1250            |
| BlueSky   | 300             |
| X         | 270             |

Replies that fail to produce usable text — empty agent responses, malformed JSON, agent errors — are dropped before they reach the platform. No empty replies. No broken formatting. No character-limit failures.

## What You Get

* Plain-text replies on every platform — no markdown leaking into Instagram comments or LinkedIn posts.
* Per-platform character limits enforced before the reply is posted.
* Empty or malformed agent responses dropped before they post — the brand never publishes a blank reply or a JSON fragment.
* Predictable formatting behaviour — same agent text rendered differently per platform, all by deterministic code.