How I Automate Content Publishing Across 10+ Platforms (Open-Source Toolkit)

I Open-Sourced a Toolkit That Automates Content Publishing Across 10+ Platforms

Managing multiple content platforms as a solo creator is exhausting. I know — I run Kuaishou, WeChat, Zhihu, Juejin, Bilibili, and several international platforms at the same time.

The biggest time sink isn't writing. It's the busywork: collecting trends, deciding what to write, adapting content per platform, running quality checks, publishing, and verifying posts actually went live.

So I built ai-self-media-tools — an open-source workflow toolkit that automates this entire loop.

What It Does

The toolkit chains the full content lifecycle into one pipeline:

1. Trend collection — pulls trending topics from multiple sources (GitHub, HN, Douyin, Bilibili, WeChat search) in one command
2. Topic scoring — ranks candidate topics by trend heat, utility, visual promise, platform fit, and historical feedback
3. Content generation — generates platform-adapted content (long-form, short-form, carousel scripts, video scripts)
4. Quality gates — blocks content that fails checks: duplicates against history, platform format rules, image relevance, license compliance, publish health
5. Draft-first publishing — pushes to platform draft boxes by default; humans review before anything goes live
6. Postcheck verification — re-checks the platform management page to confirm content actually landed (uploader "success" is not trusted)

Why Draft-First?

AI can do 90% of the work, but the final 10% — the judgment call — should stay human. The toolkit defaults to pushing content into draft boxes rather than auto-publishing. You review, you publish.

This also keeps you compliant with platforms that require human review before public posts.

Agent-Agnostic Design

It doesn't bind to one AI assistant. It runs with:

• Hermes (my daily driver)
• Codex
• Claude Code
• Or plain CLI commands

You keep your favorite AI. The toolkit provides the workflow skeleton.

Architecture

content_platform/    # Python workflow engine
skills/              # Reusable style & prompt rules
tests/               # Regression & behavior coverage
systemd/             # Deployment templates
scripts/install.py   # Cross-platform installer
docs/                # Documentation

Key engineering principles baked in:

• Clean publishable rule: a project-audit command scans for secrets (keys, cookies, IPs) before any git push
• Gate-based quality: quality gates are hard checks, not suggestions — no bypass path
• Privacy first: no credentials or cookies live in the repository

Quick Start

python scripts/install.py
python -m content_platform health
python -m content_platform trends --limit 5
python -m content_platform analyze-topic --topic "AI workflows"
python -m content_platform project-audit

Install to first trend pull: about 15 minutes.

What It Saved Me

Running this daily for my own channels (Kuaishou, WeChat, Zhihu, Juejin):

• ~2 hours/day saved on trend scanning, format adaptation, and publish verification
• Fewer duplicate-topic mistakes — the history check catches them before generation
• No more "thought I published but it was still in drafts" — postcheck catches it

Who It's For

• Solo creators managing multiple platforms (the core use case)
• Technical bloggers comfortable with CLI
• Content teams wanting a consistent quality baseline
• Data-driven creators who want trend + score + review loops

Try It

It's completely open source:

https://github.com/mage0535/ai-self-media-tools

If it saves you time too, star the repo — it helps more creators find it. Issues and PRs welcome.

Content creation is a marathon. Tools that remove repetitive work let you spend your energy where it matters: the content itself.

Which part of your multi-platform workflow eats the most time? Let me know in the comments.