Flyto2 Blog

Plain-language guides for AI workflow automation, MCP server automation, CTEM, and practical Flyto2 engineering.

Read online · Website · Docs · Open source

---

What lives here

Long-form writing that answers the questions people actually search for before they read the docs. Flyto2 Blog is the public editorial surface for AI workflow automation tools, open-source AI agent frameworks, MCP server automation, continuous threat exposure management, attack surface management, browser automation, no-code API integration, email parsing automation, SEO/GEO/AEO, and engineering release notes.

The voice should be practical and conversational: explain the problem, show the tradeoff, give a concrete Flyto2 path, then link to the docs only when the reader is ready to implement.

• Release stories — "v2.x shipped; here's what changed and why"
• Engineering deep dives — design decisions, migration stories
• Showcase posts — real automations customers have built
• SEO/AEO/GEO guides — AI search visibility, crawler logs, and answer-shaped technical content
• Long-tail explainers — "what is AI workflow automation", "open source AI agent framework", "MCP server automation", "n8n alternative", "Zapier alternative", "Make alternative", "Playwright alternative", "LangGraph alternative", and comparison-style posts
• Product-line education — automation, security, Data, Zero-person Company Agent, and Big Data / Intelligence workflows
• Post-mortems — incident reports with blameless root-cause
• Community spotlights — plugin and template authors

Format

posts/
└── stable-kebab-slug.md    # frontmatter + markdown body

Frontmatter schema lives in POSTING.md (or CONTRIBUTING.md if you're writing a post).

Architecture

VitePress renders posts/ into the public article surface. Repository scripts validate frontmatter and links, generate discovery feeds, score built SEO output, prepare reviewed social plans, and render video assets. The generated source and content reference is the method-level index; the technical whitepaper explains the editorial, search, social, and video boundaries.

API / Content Model

Each post is a Markdown file with frontmatter metadata for title, summary, date, category, tags, canonical URL, and publication status. Treat posts as a public content API: landing pages, docs, social previews, AI answer engines, and search crawlers all depend on stable titles, summaries, canonical links, and long-tail keyword phrasing.

Configuration

Keep public links canonical:

• Product pages use https://flyto2.com.
• Technical docs use https://docs.flyto2.com.
• Blog articles use https://blog.flyto2.com.
• Security contact uses [email protected].

Install

npm install

Usage

npm run dev        # http://localhost:3000
npm run build

Testing

Run the production build before publishing public content:

npm run verify

npm run verify includes a Rank Math-style SEO score gate after the static build. It scores every built page from 0-100 across technical metadata, focus keyword placement, content structure, links/images, and AI visibility signals. Reports are written to .seo/reports/seo-score.json and .seo/reports/seo-score.md. CI fails when the homepage, any post, or the site average falls below the configured thresholds.

The verification loop begins with npm run docs:check, which requires every maintained executable source, article, workflow, configuration input, distribution plan, and public asset to have a documentation owner and rejects stale generated references. Regenerate with npm run docs:reference.

npm run verify also runs npm run seo:manage. That management pass creates .seo/reports/seo-management.json and .seo/reports/seo-management.md with rank targets, keyword gaps, internal-link suggestions, Search Console opportunities when CSV exports are present, and editor recommendations for the next content pass. Raw Search Console CSV files stay local under .seo/search-console/ and are ignored by Git.

npm run build runs npm run seo:discovery before VitePress. That generator keeps public discovery surfaces in sync with Markdown posts: rss.xml, atom.xml, feed.json, image-sitemap.xml, discovery-manifest.json, and a real og-image.png fallback for social previews. npm run audit:seo fails if these files are missing, stale, or not advertised from page metadata and robots.txt.

For social drafts, run:

npm run social:dry-run

The Social Publish GitHub Action can publish approved plans to LinkedIn and a Facebook Page once the maintainer-owned secrets are configured. Without those secrets, the action records a dry-run artifact instead of failing or posting.

For YouTube/video drafts, run:

npm run video:from-post -- posts/community-growth-open-source-ai-workflow-automation.md --out video/plans/community-growth-open-source-ai-workflow-automation.json --force
npm run video:check
npm run video:qa
npm run video:storyboard

The manual Video Render GitHub Action installs render tools, turns a reviewed plan under video/plans/ into YouTube 16:9, Shorts 9:16, and LinkedIn 1:1 MP4 artifacts, and never reads YouTube tokens. Use its variant input to render one output instead of all outputs.

Contributing a post

1. Fork + branch.
2. Add posts/your-stable-kebab-slug.md; keep the publication date in frontmatter so the public URL does not change when editorial dates change.
3. If the post should be promoted, add or update a plan under social/posts/ and verify it with npm run social:check.
4. If the post should become a video, generate or update a plan under video/plans/ and verify it with npm run video:qa.
5. Open PR; tag @flytohub/editors for review.

Not a team member but want to contribute a guest post? Open a Discussion with an outline first.

Related

• flyto2.com
• flyto-docs — product documentation (not essays)
• flyto-landing-page
• flyto-indexer — code intelligence MCP for AI-assisted engineering
• Flyto2 community hub

Documentation

• Technical and editorial whitepaper
• Generated source and content reference
• Posting contract
• Social distribution
• Video production

License And Reuse

This repository does not currently declare one blanket license for editorial copy, Flyto2 trademarks, screenshots, generated media, and third-party assets. Each stock or image asset keeps the provenance recorded in its credits or catalog entry. Contact [email protected] before reusing material whose license is not explicit.