The latest articles on DEV Community by Maurizio (@anhonestboy). https://dev.to/anhonestboy https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3957376%2Fe441cbb5-34c1-4bdd-a6bc-b3527862fda1.jpeg https://dev.to/anhonestboy en Maurizio Thu, 28 May 2026 22:33:46 +0000 https://dev.to/anhonestboy/im-a-photographer-i-built-a-dsl-for-multi-agent-workflows-roast-me-k4 https://dev.to/anhonestboy/im-a-photographer-i-built-a-dsl-for-multi-agent-workflows-roast-me-k4 <p>I'm a wedding photographer. Over the past couple of months I fell down a rabbit hole and built something I honestly can't evaluate, because I don't fully understand what I built. I need your take.</p> <h2> What it does </h2> <p><strong>AgentFlow DSL</strong> — you write a <code>.aflow</code> file describing agents, phases, and loops. It auto-exposes as an MCP tool in Claude Code. Zero Python glue.</p> <p>Example: a 3-agent workflow with a quality loop:<br> </p> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code>workflow code_quality agents: agent writer → model: "local-fast" agent tester → model: "openrouter-smart" agent critic → model: "claude-sonnet" loop quality_gate phases: [write, test, review] repeat_while: review.verdict == "needs_work" max_iterations: 10 </code></pre> </div> <p>Writer generates code → Tester tries to break it → Critic reviews. If the critic says "needs_work", it loops back with feedback. 68 lines total.</p> <h2> Does it actually work? </h2> <p>Yes, somehow. Tested with real OpenRouter calls (Gemini Flash). 92 tests pass. Three executors: Claude SDK, OpenRouter API, Ollama. MIT license.</p> <p>But here's the thing: <strong>I coded most of this without fully understanding what I was doing.</strong> The tokenizer, the parser, the compiler. I wrote them, but if you asked me to explain the theory, I'd struggle. Claude Code (the irony) helped me write a lot of it. I'm genuinely not sure if I built a real compiler or just a fancy string processor with a loop.</p> <h2> My honest questions </h2> <ol> <li>Does this solve a real problem or did I reinvent Python with extra characters?</li> <li>LangGraph, CrewAI, AutoGen already exist, is a declarative DSL actually <em>better</em>, or do developers prefer writing code?</li> <li>Is "docker-compose for AI agents" a useful framing or marketing fluff?</li> <li>If you look at the code, is it architecturally solid or fundamentally wrong?</li> </ol> <h2> Links </h2> <ul> <li> <strong>Repo:</strong> <a href="https://github.com/anhonestboy/agentflow" rel="noopener noreferrer">github.com/anhonestboy/agentflow</a> </li> <li> <strong>npm:</strong> <code>@anhonestboy/agentflow</code> </li> </ul> <p><strong>I'm a photographer, not a professional developer. Tear it apart.</strong> If it's trash, say so. If there's something here, tell me what to focus on.</p> ai programming agents learning