Harrison Schaen

Drupal's robust access control system allows you to manage content creation, editing, and deletion permissions for specific roles. But what if you need to control access to individual fields within content types? Enter the Field Permissions module. This powerful tool enables granular control over field-level access, perfect for scenarios where you want to: 1️⃣ Restrict editing of specific fields to certain roles 2️⃣ Limit viewing of premium content fields to specific user groups Implementation is straightforward: 1️⃣ Download and install the Field Permissions module 2️⃣ Edit a field in your content type (or any fieldable entity) 3️⃣ Look for "Field visibility and permissions" 4️⃣ Choose from three options: ➜➜ Not set: Module doesn't control access ➜➜ Private: Only author and admins can edit/view ➜➜ Custom permissions: Specify detailed access controls For a comprehensive overview of field permissions, navigate to Reports > Field list > Permissions. This report helps identify and troubleshoot permission-related issues. By leveraging the Field Permissions module, you can fine-tune your Drupal site's access control, ensuring the right users have the appropriate level of access to your content's fields. #drupal #fields #access ~~~ If you found this post helpful. ♻️ Please repost and follow. 📧 Join my newsletter; link in the first comment.

9

1 Comment

🎉We’ve officially joined the Drupal AI Makers community because we believe AI’s true power lies in amplifying human expertise, not replacing it. Most marketers still think of Drupal as “a CMS for developers.” But Drupal’s new AI-focus is about something bigger. ➡️ We’re reinventing the strategic role of your content management system: When you combine deep audience research, well-crafted brand guidelines, and strategic content planning with AI capabilities, you give small marketing teams the superpower to match the quality and output of much larger ones. 💪 Here’s why Drupal AI Makers matters to teams that we collaborate with: ��� ✅ Small marketing teams should be able to ship campaigns fast - not get stuck waiting on bandwidth or dev cycles. ✅ Higher-ed institutions should be able to maintain brand consistency across complex ecosystems - without sacrificing governance. We’re excited to bring our perspective to this project, making AI in Drupal truly marketer-focused. #Drupal #DrupalAIMakers #AIForMarketers #MarketingInnovation #OpenSource

56

2 Comments

Level up your Claude Code game. 9 steps from beginner to expert. Most people maybe write a few lines of context for Claude, and stop there. These nine steps are how I turned Claude Code into the operating system for my agency. STARTER (do these today) 1. Stop re-explaining yourself. CLAUDE.md is a file Claude reads every session. Write your rules once in plain English. I wrote "UK English, no jargon, short sentences." Every email, proposal, and draft has followed that since day one. 2. Kill the same mistakes permanently. Add a forbidden section listing words, phrases, and patterns it must never use. Finally, get rid of those em dashes (sorry, grammarians, I used to love these too). 3. Never repeat a correction. Say "/meta-rule" and describe what went wrong. Claude writes it to memory and applies it to every future session. Claude once hallucinated a statistic. One command fixed that permanently. INTERMEDIATE (compound over weeks) 4. Make every output match your standards. Point Claude at your existing documents and tell it when to read them. My voice guide is 587 lines. Claude reads it before drafting anything. The output sounds like me, not like AI. 5. Pick up where you left off. Session notes in a dedicated folder tell the next conversation what happened in the last one. I ran a 14-post content review across three sessions. Each one started at exactly the right post with all decisions intact. 6. Quality gates you don't have to remember. Hook scripts in .claude/hooks/ validate actions in the background. I have one hook that captures context from every git commit and another that checks credentials before external calls. They run silently. I don't think about them. ADVANCED (build once, use forever) 7. One sentence, full execution. Slash commands in .claude/commands/ map a trigger to a multi-step workflow. "/meeting" triggers contact classification, CRM record, calendar invite, prep brief, and meeting note. Five systems updated from one line. 8. Get better without trying. A cumulative learnings file grows with every session. Claude reads it before the next one. Every editorial correction gets recorded and applied next time. My approval rate climbs without repeating myself. 9. Stop copy-pasting IDs and references. One YAML config file holds every Notion database, Google Sheet, and workflow ID in my system. Every script reads from it. When I add a new integration, I update one file. Nothing is hardcoded. I started with the first three. Now it's the operating system for my consulting practice: 350 lines of rules, 24 commands, 30 workflows. All self-taught. All plain English. Claude rewards experimentation. You just need to start. The full cheatsheet covers all nine steps with templates and examples. Grab it here: https://lnkd.in/ez2WbR4b ♻️ Repost to help someone in your network level up with Claude Code. 🔔 Follow Tom Jones for more on AI, operations, and leadership.

30

26 Comments

Yes, the best way to show up in the new world of search is to revamp your information architecture. But I'm seeing a lot of very sketchy content structure and navigation design recommendations coming from the SEO influencer hive. Their solutions for "query fan out" are.... middling, at best. No, you don't need "fraggles" or "chunking" or whatever term they've invented related to 1980s kids' entertainment. The concepts behind these recommendations are worth your attention, but the methods SEO leaders are advocating to your UX and info architecture are... less than ideal. They've nailed the cause, but not the cure. "Chunking" is trying to reverse engineer what UX folks call Content Modeling. Content Modeling is an established concept in content strategy that identifies core components of content and creates structured interfaces like SERPs. Or streaming platforms. Or AI overviews. (I learned it in 2014.) Another, similar approach is called object-oriented UX (OOUX). Both OOUX and Content Modeling are top-down structural evaluations based on the neuroscience of how people process and evaluate content. When people like and use your content because it's well-structured, it becomes mentioned and recommended more often, builds authority, EEAT, etc. (And yes, I have proof.) If you value your audience: don't change your navigation and page linking strategy without consulting someone who knows UX. SEO folks have a long trail of web garbage behind them, especially in structure and navigation. I saw a prominent influencer recommend "more in-line links" for "better UX" and... clearly he has never read the many pieces of research that say in-line links make navigation more confusing for all but the heaviest internet users. Navigation and information architecture belong with systems thinkers and editorial organizers who understand big-picture content strategy. I'm slotting some tips on info architecture optimization into my newsletter for the next month. (And the newsletter resumes tomorrowI I promise!) And in the meantime: - The great Sophia V Prater has many OOUX resources to share. Google "OOUX," and she owns it. - Button is an upcoming virtual conference about Content Design from the legendary creators of aforementioned Confab. - Greg Dunlap recently published a book called Designing Content Authoring Experiences, if you're looking for a more advanced look at creating systems that creators can use to build better structure. If you've never taken a content design or UX course before, the above are great places to start, before you go tearing up your navigation because some dude thinks you'll be mentioned in LLMs with the chunking. And if you need support on an upcoming IA project (especially prepping IA for AI), raise your hand and I'll send the names of some very smart structural thinkers.

41

10 Comments

Anthropic just shipped Agent Teams into Claude Code and it completely changes how you run creative research and campaign workflows 🤯 OpenClaw has been blowing up up. 140K+ GitHub stars in a week. Everyone’s trying to configure multi-agent setups with it right now. Custom skills, MCP integrations, environment variables, Docker containers. It works. But let's be honest: it’s unnecessarily complicated for what 99% of business owners actually need. Anthropic saw the demand and made it native. No Mac mini, no VPS, not 6-hour installation setup. Instead, they built right into Claude Code. Here's how it works: Instead of one agent doing everything solo in a straight line, a lead agent breaks your task into pieces, spins up multiple teammates, and they all work on different parts of your project at the same time. One researches competitors, one audits your landing page, one analyzes ad creative. They talk to each other, share findings, and coordinate through a shared task list without you managing any of it. This is different from sub-agents. Sub-agents can only report back to the parent. They can't message each other or share discoveries mid-task. Agent teams can. That's the unlock. Best use cases I'm seeing for e-comm and agencies: → Competitor ad research across Meta, TikTok, and YouTube in parallel → Landing page QA from 4 angles at once (conversion, performance, copy, technical) → Creative brief development where audience research, competitive analysis, and concept ideation happen simultaneously → Product catalog audits where 4 agents each handle a segment of your SKUs → Campaign launch QA checking pixels, UTMs, forms, and compliance at the same time The catch: It's experimental. It burns through tokens fast. Each teammate has its own context window so usage scales with team size. But for complex marketing workflows where parallel work actually matters, it's a game changer. I put together a complete breakdown covering: → What agent teams are and how they actually work → How they're different from sub-agents (and when to use which) → How to enable them in your settings in 30 seconds → Copy-paste prompts for competitor research, landing page QA, and creative briefs → Best practices so you don't burn through tokens → The limitations you need to know before you start This is one of those features that sounds small on paper but completely changes how you run research and QA once you start using it. Want access completely for free? > Like this post > Comment "AGENT" And I'll send it over (must be following so I can DM)

68

100 Comments

Roo Code lets you mold your #AI coding agent into what you actually need. Brandon Mathis from This Dot Labs built custom modes for documentation, pair programming, writing specs, and improving his prompts. If you're serious about agent workflows (which you have to be), this is worth a read.

13

4 Comments

Google can now supply page text directly to Gemini via url_context tool. Catch 22, it comes from internal storage and not the live site. I've extracted internal tool instructions and included them in the article. https://lnkd.in/gbUJasmJ

82

8 Comments

Claude just killed our startup. That’s how Ira Bodnar started her story on X I found today. They built an AI agent that automates ad management for you. They got several hundred paying clients in 2 months, was growing like crazy. Then Claude and Manus both released connectors for Meta Ads. Close rate dropped to 20%. Nothing dramatic happened. No scandal. No mistake. Just a model update. And this is the scary part - there will be more stories like this. If your product is "AI that does X", and the main models decide to do X natively, you’re in trouble. Not because you built badly. But because the ground is moving under you. Last week I wrote that SaaS is changing. Building is cheap. Everyone can ship. Supply explodes. This is what that looks like in real life. This story proves the point. What did they do? They pivoted. Now they focus on complex workflows for large agencies. They offer AI-powered services to SMBs. They sell outcomes, not just software access. This is the new "Service-as-Software" model. That’s the shift. Generic SaaS is fragile. Execution is harder to copy. Trust is harder to automate. Distribution is harder than ever. I think we will see more "Claude/OpenAI killed our startup" posts in 2026.

41

10 Comments

Gracefully Handled the time/Date mismatch in Revlead without any package or plugin. Since the app was depending a lot on the DateTime.now() (Device Time). Thats where the features are disturbed and data is disrupted. How i did it. 1. Deployed supabase edge function returning server time,. 2. Converted Server Time and Device Time to UTC. 3. Find the time difference and set the error margin (in minutes). 4. If difference is greater than margin then time is Incorrect. #flutter #dart

13

An AI agent just rebuilt a 450+ page website in one shot. Link below. Content, layout, media. No manual work. It’s the future for websites. The old way was CMSs, plugins, page editors, and constant upkeep. The new way is simple: code plus intent. Content becomes structured data. Design becomes rules. An agent handles builds, edits, fixes, and upgrades. You don’t click pages or collections. You say what needs to change so that the system does the rest. Websites won’t be edited anymore. They’ll be run.

28

1 Comment

Right now everyone is trying to crack AI static ads the same way. JSON prompts. 4 different LLMs. Copy-pasting between tools. Tweaking parameters. Debugging hands. Fixing garbled text. Spending 45 minutes on one ad that still needs a designer to clean up. I know because I built that exact workflow. I shared it publicly. 859 people wanted it. And then I stopped using it. Here is what most people are doing right now: 1. Write a detailed JSON prompt with product specs, persona, angle, layout. 2. Feed it to GPT or Gemini for the image. 3. Realize the text is wrong. 4. Open Canva or Figma to fix it. 5. Try again with a different model. 6. Repeat 10 more times. 6 steps. 4 tools. 45 minutes. One static ad. Here is what I do now: Pick a product. Pick a theme. Pick an angle. Click generate. Launch. 3 clicks. Done. No JSON. No prompt engineering. No bouncing between tools. The strategic thinking is already built into the system. Hooks. Layouts. Visual hierarchy. What actually converts on Meta. Trained on what we have learned managing millions in ad spend across 150+ brands. This is the difference between engineering a prompt and engineering a system. One makes you feel productive. The other makes you profitable. I put together a breakdown showing exactly how the old prompt workflow compares to the 3-click system we use now. Comment "STATIC" and I will send it over.

165

452 Comments

Claude + Airtable MCP + Wispr  The holy trinity for modern content creation. Here's how I use it: Build an Airtable database with the best inspo content Connect it to Claude via the Airtable MCP Make sure you're using Claude Desktop Use Wispr to speak your ideas to Claude Claude writes the first draft Use your human brain to take it the rest of the way Download those 3 tools and try this system for yourself. And follow me Jake Abrams as I write about AI & advertising.

48

17 Comments

We're building the ultimate website auditing system at Ahrefs. -IndexNow integration - tells us when things are updated + submits to search engines -Patches (JavaScript & now Cloudflare workers) - automated fixes of issues -Always On Audit - constantly crawling + notifications of issues We need to hook up the IndexNow data to prioritize Always On Audit + build out some more use cases for Patches, but this dream setup is coming together.

105

20 Comments