--- title: Specification-driven — The Agentic AI Hub | daily.dev description: You define what to build instead of building it: write the spec, set the tests, judge the results. Leave for twelve hours and come back to a green suite. canonical: https://daily.dev/agentic-ai-hub/paths/l4/ og:type: website og:url: https://daily.dev/agentic-ai-hub/paths/l4/ og:title: Specification-driven — The Agentic AI Hub | daily.dev og:description: You define what to build instead of building it: write the spec, set the tests, judge the results. Leave for twelve hours and come back to a green suite. og:image: https://daily.dev/assets/og/agentic-ai-hub-l4.png?v=a95f22fb og:site_name: daily.dev og:locale: en_US twitter:card: summary_large_image twitter:site: @dailydotdev twitter:creator: @dailydotdev twitter:title: Specification-driven — The Agentic AI Hub | daily.dev twitter:description: You define what to build instead of building it: write the spec, set the tests, judge the results. Leave for twelve hours and come back to a green suite. twitter:image: https://daily.dev/assets/og/agentic-ai-hub-l4.png?v=a95f22fb --- [Skip to main content ](#main-content) L4 # L4: Specification-driven You define what to build instead of building it: write the spec, set the tests, judge the results. Leave for twelve hours and come back to a green suite. [Start reading → ](/agentic-ai-hub/agentic-engineering-core-ideas/?path=l4&i=0#the-harness-as-a-first-class-artifact) [Check yourself first ↓ ](#quiz) 1. [ 1 The harness as an artifact In Agentic Engineering: Core Ideas "Agentic coding" is an overloaded term. › ](/agentic-ai-hub/agentic-engineering-core-ideas/?path=l4&i=0#the-harness-as-a-first-class-artifact) 2. [ 2 Evals-as-tests In Verification & Testing for Agents If generation is cheap and verification is the bottleneck, then verification infrastructure is your leverage. › ](/agentic-ai-hub/verification-testing-for-agents/?path=l4&i=1#evals-as-tests) 3. [ 3 LLMOps & guardrails Shipping LLM features without evals and observability is flying blind: outputs are non-deterministic, quality is subjective, and regressions are silent. › ](/agentic-ai-hub/llmops-evals-observability-guardrails/?path=l4&i=2) ## Check yourself. One question per step. Take it cold to find where to start, or after reading to see what stuck. Nobody's grading you. Best so far: Question 1 of 3 1. Step 1 Your agent keeps failing on a task. Where does the handbook say to look first? A The model: wait for a smarter release B The harness: prompts, tools, hooks, context strategy C The task: it may be beyond agents entirely D The budget: it needs more tokens per run Most agent failures are harness gaps, not model gaps. A decent model with a great harness beats a great model with a bad one. [Covered in step 1 · The harness as an artifact → ](/agentic-ai-hub/agentic-engineering-core-ideas/?path=l4&i=0#the-harness-as-a-first-class-artifact) Next question → 2. Step 2 Your test suite is green. Why run evals on agent behavior too? A Evals are cheaper to run than unit tests B Evals remove the need for human review C Regulators increasingly require behavioral evals D Tests check the product, not the process that produced it A prompt, model, or harness change can silently degrade agent behavior while every existing test stays green; evals-as-tests catch those regressions. [Covered in step 2 · Evals-as-tests → ](/agentic-ai-hub/verification-testing-for-agents/?path=l4&i=1#evals-as-tests) Next question → 3. Step 3 How should prompts be managed once a feature is in production? A Versioned like code, with evals gating each change B Hard-coded so they can't drift from the app C Edited in place for the fastest iteration D Frozen after launch to avoid regressions Version prompts like code: store them outside the app, track versions, run evals per version, and roll back. The anti-pattern is prompts hard-coded and edited in place with no history or eval gate. [Covered in step 3 · LLMOps & guardrails → ](/agentic-ai-hub/llmops-evals-observability-guardrails/?path=l4&i=2) See results → of 3 * [ 1 The harness as an artifact → ](/agentic-ai-hub/agentic-engineering-core-ideas/?path=l4&i=0#the-harness-as-a-first-class-artifact) * [ 2 Evals-as-tests → ](/agentic-ai-hub/verification-testing-for-agents/?path=l4&i=1#evals-as-tests) * [ 3 LLMOps & guardrails → ](/agentic-ai-hub/llmops-evals-observability-guardrails/?path=l4&i=2) Try again [Where to next → ](#path-next-title) ## Where to next. Climb to the next rung, or see how the other levels work. 1. L0 ### L0: Manual The ground floor: AI as a search engine and occasional tab-complete. No real productivity change. Everyone starts here. There's nothing to unlock yet. 2. L1 ### L1: Discrete task offloading You hand the AI small, self-contained jobs: write the unit tests, draft the docstring, generate one well-scoped function. The real work is still yours. Chat assistants · Prompting 101 · IDE assistants [Explore path → ](/agentic-ai-hub/paths/l1/) 3. L2 ### L2: Active pairing The AI writes alongside you all day, handling the boring parts while you steer and review everything as it lands. Shapiro estimates about 90% of AI-native developers live here. CLI agents · AGENTS.md & CLAUDE.md · What an agent actually is [Explore path → ](/agentic-ai-hub/paths/l2/) 4. L3 ### L3: Human-in-the-loop management You mostly stop typing code. Agents run several tasks at once and your day becomes reviewing their work. Most people never go past this level. Own the outer loop · Agentic code review · Role separation · MCP & tools [Explore path → ](/agentic-ai-hub/paths/l3/) 5. L4 ### L4: Specification-driven You define what to build instead of building it: write the spec, set the tests, judge the results. Leave for twelve hours and come back to a green suite. The harness as an artifact · Evals-as-tests · LLMOps & guardrails You're on this rung 6. L5 ### L5: The dark factory Agents building the software with humans out of the loop. Shapiro reports it working only for teams under five people, and calls it likely our future. Swarms & fleets · The factory model · Human factors [Explore path → ](/agentic-ai-hub/paths/l5/) [All reading paths → ](/agentic-ai-hub/#paths-start-here) [Browse every chapter ](/agentic-ai-hub/#pages) ```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/og-image.png?v=a830cdf1","width":1200,"height":630},"sameAs":["https://twitter.com/dailydotdev","https://www.linkedin.com/company/dailydotdev","https://github.com/dailydotdev","https://www.instagram.com/dailydotdev"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","description":"Free, personalized developer news aggregator. Stay on top of software development news, AI coding tools, and web dev - curated daily from trusted sources.","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"}},{"@type":"WebPage","@id":"https://daily.dev/agentic-ai-hub/paths/l4/","url":"https://daily.dev/agentic-ai-hub/paths/l4/","name":"Specification-driven | The Agentic AI Hub","description":"You define what to build instead of building it: write the spec, set the tests, judge the results. Leave for twelve hours and come back to a green suite.","inLanguage":"en-US","isPartOf":{"@id":"https://daily.dev/#website"}},{"@type":"ItemList","numberOfItems":3,"itemListElement":[{"@type":"ListItem","position":1,"url":"https://daily.dev/agentic-ai-hub/agentic-engineering-core-ideas/#the-harness-as-a-first-class-artifact","name":"The harness as an artifact"},{"@type":"ListItem","position":2,"url":"https://daily.dev/agentic-ai-hub/verification-testing-for-agents/#evals-as-tests","name":"Evals-as-tests"},{"@type":"ListItem","position":3,"url":"https://daily.dev/agentic-ai-hub/llmops-evals-observability-guardrails/","name":"LLMOps & guardrails"}],"@id":"https://daily.dev/agentic-ai-hub/paths/l4/#steps"},{"@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev/"},{"@type":"ListItem","position":2,"name":"The Agentic AI Hub","item":"https://daily.dev/agentic-ai-hub/"},{"@type":"ListItem","position":3,"name":"The adoption ladder","item":"https://daily.dev/agentic-ai-hub/#levels"},{"@type":"ListItem","position":4,"name":"Specification-driven","item":"https://daily.dev/agentic-ai-hub/paths/l4/"}]}]} ```