--- title: AI Assistant for Our Blog Writing Process | daily.dev description: A developer at OmbuLabs built a blog writing assistant using an MCP server and a Claude Code plugin to address quality and consistency problems introduced by... canonical: https://daily.dev/posts/ai-assistant-for-our-blog-writing-process-uduobqhmo twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: AI Assistant for Our Blog Writing Process | daily.dev og:description: A developer at OmbuLabs built a blog writing assistant using an MCP server and a Claude Code plugin to address quality and consistency problems introduced by... og:url: https://daily.dev/posts/ai-assistant-for-our-blog-writing-process-uduobqhmo og:image: https://api.daily.dev/og/posts/UDUOBQHmo.png og:image:alt: AI Assistant for Our Blog Writing Process og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # AI Assistant for Our Blog Writing Process **[RUBYLAND](https://daily.dev/sources/rubyla)** · 12 min read · 1 upvotes · 0 comments ## Summary A developer at OmbuLabs built a blog writing assistant using an MCP server and a Claude Code plugin to address quality and consistency problems introduced by AI-assisted writing. The solution includes a data pipeline that pulls markdown posts from GitHub into a Postgres database with pgvector embeddings, an MCP server exposing blog taxonomy and style-check tools, and a Claude Code plugin packaging the team's writing workflows. The whole system took about ten hours to build, cost roughly $2,000 in time, and is expected to cut review time by 30–40%. The key insight is that a constrained, data-aware extension of an existing AI tool was far more practical than building a custom agent from scratch. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can I use pgvector with Postgres to find related blog posts by meaning instead of keywords? Store an embedding, a numeric summary of each post's content, in Postgres using the pgvector extension. Posts covering similar ground end up with similar embedding vectors, which lets an application query for semantically related content rather than relying on exact keyword matches. This approach was used to power topic and related-post suggestions inside an AI writing assistant. _Developers comparing vector search approaches with pgvector can track real implementation patterns on daily.dev._ ### What is the difference between building a custom AI agent and using an MCP server plus a Claude Code plugin for a workflow tool? A custom agent requires a hosting environment, ongoing maintenance, and significant upfront engineering for orchestration and tool access, while an MCP server plus a Claude Code plugin reuses an existing agent (Claude Code) and only adds domain-specific data access and packaged workflows. One team built the MCP server and plugin combination in about ten hours and roughly $2,000, versus an estimated weeks and tens of thousands of dollars for a custom agent doing the same job. _Anyone weighing custom agents against lighter MCP-based tooling can follow real cost comparisons on daily.dev._ ### Why does using AI to draft content create more work for editors and reviewers instead of less? AI tends to fill gaps with assumptions and rewrite carefully hedged claims into cleaner but inaccurate statements, which shifts the burden from writing to reviewing. In one team's case, this meant reviewers received more posts with more issues to flag, review cycles took longer, and more back-and-forth landed on authors, so the productivity gain from faster drafting turned into a net productivity decrease at the review stage. _Teams rethinking AI-assisted writing workflows can compare real trade-offs like this on daily.dev._ ## Similar posts on daily.dev - [Building an AI Blog Editor with Claude Skills](https://daily.dev/posts/building-an-ai-blog-editor-with-claude-skills-wlnarm7rp) · RUBYLAND · 0 upvotes · 0 comments - [Building a Blog With AI](https://daily.dev/posts/building-a-blog-with-ai-b8x6jproz) · Viget · 1 upvotes · 0 comments - [WordPress.com now lets AI agents write and publish posts, and more](https://daily.dev/posts/wordpress-com-now-lets-ai-agents-write-and-publish-posts-and-more-mlj8daq2v) · TechCrunch · 2 upvotes · 0 comments - [How AI made my writing process faster without lowering the bar](https://daily.dev/posts/how-ai-made-my-writing-process-faster-without-lowering-the-bar-ty0pt7gia) · We Are .NET · 0 upvotes · 0 comments - [How I Built a Blog-to-Podcast AI Agent](https://daily.dev/posts/how-i-built-a-blog-to-podcast-ai-agent-6hfveoxdq) · Medium · 2 upvotes · 0 comments --- Tags: [#devtools](https://daily.dev/tags/devtools), [#mcp](https://daily.dev/tags/mcp), [#claude-code](https://daily.dev/tags/claude-code), [#pgvector](https://daily.dev/tags/pgvector) [View this post on daily.dev](https://daily.dev/posts/ai-assistant-for-our-blog-writing-process-uduobqhmo) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"AI Assistant for Our Blog Writing Process","url":"https://daily.dev/posts/ai-assistant-for-our-blog-writing-process-uduobqhmo","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/ai-assistant-for-our-blog-writing-process-uduobqhmo"},"datePublished":"2026-07-30T19:25:32.021Z","dateModified":"2026-09-14T07:04:38.774Z","description":"A developer at OmbuLabs built a blog writing assistant using an MCP server and a Claude Code plugin to address quality and consistency problems introduced by...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e1b2f6d8b65be103000eb3546edaac5a?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e1b2f6d8b65be103000eb3546edaac5a?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"RUBYLAND","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"RUBYLAND","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/rubyla","url":"https://daily.dev/sources/rubyla"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/ai-assistant-for-our-blog-writing-process-uduobqhmo","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"devtools,mcp,claude-code,pgvector","timeRequired":"PT12M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"RUBYLAND","item":"https://daily.dev/sources/rubyla"},{"@type":"ListItem","position":3,"name":"AI Assistant for Our Blog Writing Process"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/ai-assistant-for-our-blog-writing-process-uduobqhmo#faq","mainEntity":[{"@type":"Question","name":"How can I use pgvector with Postgres to find related blog posts by meaning instead of keywords?","acceptedAnswer":{"@type":"Answer","text":"Store an embedding, a numeric summary of each post's content, in Postgres using the pgvector extension. Posts covering similar ground end up with similar embedding vectors, which lets an application query for semantically related content rather than relying on exact keyword matches. This approach was used to power topic and related-post suggestions inside an AI writing assistant. Developers comparing vector search approaches with pgvector can track real implementation patterns on daily.dev."}},{"@type":"Question","name":"What is the difference between building a custom AI agent and using an MCP server plus a Claude Code plugin for a workflow tool?","acceptedAnswer":{"@type":"Answer","text":"A custom agent requires a hosting environment, ongoing maintenance, and significant upfront engineering for orchestration and tool access, while an MCP server plus a Claude Code plugin reuses an existing agent (Claude Code) and only adds domain-specific data access and packaged workflows. One team built the MCP server and plugin combination in about ten hours and roughly $2,000, versus an estimated weeks and tens of thousands of dollars for a custom agent doing the same job. Anyone weighing custom agents against lighter MCP-based tooling can follow real cost comparisons on daily.dev."}},{"@type":"Question","name":"Why does using AI to draft content create more work for editors and reviewers instead of less?","acceptedAnswer":{"@type":"Answer","text":"AI tends to fill gaps with assumptions and rewrite carefully hedged claims into cleaner but inaccurate statements, which shifts the burden from writing to reviewing. In one team's case, this meant reviewers received more posts with more issues to flag, review cycles took longer, and more back-and-forth landed on authors, so the productivity gain from faster drafting turned into a net productivity decrease at the review stage. Teams rethinking AI-assisted writing workflows can compare real trade-offs like this on daily.dev."}}]} ```