Most MCP servers never get indexed anywhere that matters. The official MCP Registry, Smithery, Glama, PulseMCP — each has its own submission path, and submitting to all of them once is an afternoon you didn't plan to spend. Staying visible after that is a recurring job nobody signs up for.
I built marketing-pipeline to handle the recurring work. One command onboards a project:
marketing onboard --name my-tool --repo owner/repo --kind mcp-server
That fetches the README, sends it to Claude, and writes problem statements, factual claims, and rotation angles to projects.yml. From that config, the pipeline knows to submit to the MCP Registry (which Glama and PulseMCP pull from automatically), Smithery, and GitHub Topics — and to draft posts for Bluesky, Dev.to, Hashnode, and Mastodon.
The daily cron runs via GitHub Actions at 14:00 UTC on weekdays, rotating through every registered project × angle × channel combination, always picking the least-recently-used angle so nothing repeats in a short window.
Quality control is enforced by a hard-reject gate in pipeline/antislop.py before anything posts: 'excited', 'game-changer', 'AI-powered', hashtags, emoji, and exclamation points are blocked at publish time, not just during drafting. Per-channel length limits are also enforced — 300 chars for Bluesky, 280 for X, 150–400 words for Dev.to and Hashnode.
One constraint worth knowing: awesome-claude-code cannot be automated. Their rules require a human to submit via their GitHub issue form, so the pipeline generates the payload and stops there.
You need an Anthropic API key and credentials for at least Bluesky, Dev.to, and Hashnode to run it. Mastodon and Slack are optional.
github.com/robertnowell/marketing-pipeline
Top comments (0)