blog-seo
Self-improving technical blog engine
Watch an AI agent learn to write technical content that ranks.
About this project
Most technical blogs are written once and forgotten. blog-seo creates a self-improving loop that writes code-heavy tutorials, measures search performance via Google Search Console, learns what works, and generates better content each cycle.
Live Dashboard
Real-time performance data from the blog-seo project
How it works
Feedback
Collects search performance from Google Search Console. Identifies which posts rank well and which are stagnant or declining.
Learn
Analyzes high-performing posts to find patterns — what topics, structures, and code density correlate with traffic. Updates generation prompts.
Plan
Identifies topic gaps from GSC query data — queries with impressions but no matching content. Selects underperformers for rewrites.
Generate
Writes new posts or rewrites underperformers. Each is validated for structure, code correctness, quality, and SEO before publishing.
How WarpMetrics makes this possible
blog-seo uses the @warpmetrics/warp SDK to track every phase, every LLM call, and every outcome — creating the data layer that powers the improvement loop.
Run tracking
Each cycle is a tracked run with groups for Feedback, Learn, Plan, and Generate.
Outcome classification
Every run records outcomes: posts generated, rewritten, patterns learned. Classified as success or failure.
Self-chaining via Acts
Each run creates a "Continue Optimization" act linking to the next, forming an unbroken chain.
LLM observability
Every OpenAI call is traced — outline, full post, quality validation. Tokens, cost, model usage tracked.
Ask Anything
Query this project's data using natural language — powered by Claude with MCP tools
Ask anything about this project's data
Track your own agent improvement loop
Get the same visibility into your AI agents. Free to start, takes 2 minutes.