Luis Mori Guerra

A simple, source-backed guide for engineers and PMs to start with GitHub Spec Kit, understand the main commands, and use the workflow cleanly in Cursor or Claude Code.

Topics

Spec Kit Cursor Claude Code Spec-Driven Development Product Management Engineering Workflow

Read article

142

Articles published

292

Topics covered

10+

Years building

Browse everything

View all articles

Spec Kit OpenSpec Spec-Driven Development AI Coding

Extend GitHub Spec Kit: Add Explore and Alternatives-Review Modes

Spec Kit jumps straight to specify. Borrow two missing phases from OpenSpec-style workflows — an Explore mode and an Alternatives-Review mode — as two small extension commands, without forking or switching tools.

Updated Jul 20, 2026 18 min read

Spec Kit Spec-Driven Development AI Coding Developer Productivity

How to Make GitHub Spec Kit Produce Leaner Output

Spec Kit fans a small change out into eight-plus files. Four practical levers — inline prompts, lean templates, trimmed commands, and impact-based routing — make its specs, plans, and tasks shorter without breaking the workflow.

Updated Jul 20, 2026 13 min read

GitHub Spec Kit Agent Skills Spec-Driven Development Prompt Engineering

Why GitHub Spec Kit Is So Interesting: Six Skills Worth Studying

A prompt-engineering teardown of Spec Kit's six core skills—and a practical blueprint for building an opinionated company spec workflow with custom Agent Skills.

Updated Jul 20, 2026 28 min read

AI LLM AI Engineering Evaluation

Do New Frontier LLMs Really Resolve Ambiguity Better?

Frontier labs increasingly claim their models understand intent with less prompting. Here is what the evidence supports, what training changed, and how to test ask-versus-guess behavior in your own agents.

Updated Jul 17, 2026 21 min read

AI Agents AGENTS.md Agent Skills MCP

Let's Build Prompt-First Repositories for Your Applications

Modern production apps require many layers to maintain. A prompt-first repository adds instructions, skills, MCP, and bounded loops so agents can move faster without bypassing reliability controls.

Jul 17, 2026 23 min read

Claude Anthropic Fable 5 Opus 4.8

Claude Low Effort Compared: Fable 5 vs Opus 4.8 vs Sonnet 5

Compare Fable 5 Low, Opus 4.8 Low, and Sonnet 5 Low for coding and planning, including task cost, cross-effort surprises, and a practical routing policy.

Updated Jul 16, 2026 16 min read

Claude Anthropic Low Effort Model Routing

How to Implement and Evaluate a Low-Effort Model Routing Policy

Build a small Claude model router with explicit task classes, Low effort, bounded tools and budgets, isolated workspaces, acceptance tests, and cost-per-result evaluation.

Updated Jul 16, 2026 41 min read

Ultracode Agno AgentOS AI Agents

Build an Ultracode-Like Coding Workflow with Agno and AgentOS

A practical Agno-only blueprint using Workflow, native Agents, session state, bounded repair loops, persisted HITL review, and isolated coding workspaces.

Updated Jul 15, 2026 20 min read

Ultracode LangGraph LangChain AI Agents

Build an Ultracode-Like Coding Workflow with LangGraph

A practical LangGraph-centered blueprint using typed graph state, dynamic Send fan-out, PostgreSQL checkpoints, deterministic candidate testing, interrupts, and isolated coding sandboxes.

Updated Jul 15, 2026 19 min read

dbt DuckDB Snowflake Data Engineering

DuckDB in Dev, Snowflake in Prod: A Practical dbt Workflow for a Medallion Project

Fusion's DuckDB adapter can make a large dbt project feel local and fast, but it does not make DuckDB equivalent to Snowflake. Here is a practical two-engine workflow for an existing app, including data fixtures, model portability, CI gates, failure modes, and troubleshooting.

Jul 15, 2026 32 min read

Claude Anthropic Fable 5 Opus 4.8

Fable 5 High vs Opus 4.8 High vs Sonnet 5 High: A Practical Claude Model Guide

Compare Fable 5, Opus 4.8, and Sonnet 5 at High effort using CursorBench 3.2, then see when Ultracode changes the model-routing decision.

Updated Jul 15, 2026 19 min read

Claude Code Cursor Ultracode AI Agents

Claude Code Ultracode Architecture Blueprint for Cursor and Custom Agents

An architecture blueprint for translating Ultracode's high-effort planner, parallel workers, verification, state, and guardrails into Cursor or a custom agent harness.

Updated Jul 15, 2026 23 min read

OpenSpec Superpowers Spec-Driven Development AI Coding

OpenSpec vs Superpowers: Spec Generation and Implementation Compared

A side-by-side comparison of OpenSpec and Superpowers for engineers and architects: what each tool actually ships, how spec generation differs, and where implementation discipline comes from.

Updated Jul 14, 2026 15 min read

WebAssembly AI Vibe Coding Edge Computing

WebAssembly in 2026: What It Means for AI, Vibe Coding, and the Future of the Internet

A source-backed recap of WebAssembly in mid-2026 — Wasm 3.0, the Component Model, WASI 0.3, browser AI inference, and why WASM is the runtime layer behind vibe-coding sandboxes and portable edge software.

Jul 10, 2026 15 min read

Agentic Workspace Git Submodules Agent Skills OpenSpec

Agentic Workspaces: Submodules, Skills, Knowledge, and Spec-Driven Delivery

A source-backed guide to building an agentic workspace that coordinates multiple repos, shared skills, durable knowledge, and OpenSpec or Spec Kit — for engineers, architects, PMs, and scrum masters.

Jul 8, 2026 22 min read

Agentic Workspace Cloud Agents Cursor Claude Code

Running Agentic Workspaces in the Cloud at Scale

Running one workspace agent in the cloud is a trick; running ten reliably is an engineering discipline. A source-backed guide to environment-as-code, secrets, private submodule auth, fan-out topology, and the review capacity that actually limits throughput.

Jul 8, 2026 13 min read

Agentic Workspace Retrospectives Continuous Improvement AI Agents

Session Retrospectives: Turning Agent Runs into Platform Improvements

Every agent run leaves signals about what your workspace platform is missing. A practitioner guide to the retro-to-backlog pipeline: a structured template, a failure taxonomy, honest metrics, partial automation, and governance that avoids retro theater.

Jul 8, 2026 14 min read

Git Submodules Agentic Workspace Dependency Management CI/CD

Submodules as a Dependency Contract: Freshness Without Breakage

A deep, source-backed guide to treating git submodules as a versioned, reviewable dependency contract — freshness strategies, the full breakage taxonomy, an automation ladder, and what it means for AI and cloud agents.

Jul 8, 2026 15 min read

MCP Agentic Workspace Debugging Incident Response

Trace, Triage, Debug: MCP Tooling for Agentic Workspaces

A source-backed guide to composing MCP servers into triage and debug skills: the first-party server landscape, four composition patterns, least-privilege scoping, and the context-bloat tax that quietly breaks agents.

Jul 8, 2026 13 min read

Pi Coding Agents OpenClaw Agent Harness

Pi (pi.dev): The Minimal Coding Agent Harness and How It Compares

A source-checked guide to Pi — the minimal terminal coding harness behind OpenClaw — with comparisons to Claude Code, Cursor, Codex, and OpenCode, plus handoff workflows and when the minimal approach wins.

Jul 7, 2026 15 min read

Vibe Coding Cursor Claude Code OpenSpec

Advanced Vibe Coding: A Modern Agentic Workflow for Beginners

A practical guide to the full vibe-coding loop — explore with Ask mode and OpenSpec, install skills, generate with Plan mode or spec-driven tools, review with Bugbot and /simplify, verify in the browser with Playwright MCP, deploy with MCP, and capture lessons in custom skills.

Jul 6, 2026 18 min read

NATS Event-Driven Architecture Pub/Sub Microservices

Core NATS: Subjects, Pub/Sub, Request/Reply, and Queue Groups

A practical tour of Core NATS — how subjects and wildcards route messages, why delivery is at-most-once, and how request/reply and queue groups replace load balancers for many microservice patterns.

Jul 6, 2026 9 min read

Event-Driven Architecture NATS CQRS System Design

EDA with NATS: One Truth, Many Read Models — CQRS Projections

How a generic order domain flows from a relational write database through NATS KV and index subjects into OpenSearch and an analytics warehouse — without mistaking plumbing for source of record.

Updated Jul 6, 2026 11 min read

Event-Driven Architecture NATS JetStream Observability

EDA with NATS: Observability and SLOs for JetStream Projection Pipelines

What to monitor on NATS and JetStream, how to wire Prometheus and Grafana alerts, and which application-level SLOs catch projection lag before an SLO breach — using the orders/CDC/indexer pattern from Part 4.

Jul 6, 2026 16 min read

NATS JetStream Key-Value Store Event-Driven Architecture

JetStream KV Is a Change Stream, Not a Table

JetStream KV looks like a key-value database, but it is a stream abstraction with a latest-value view — watch for changes, design keys as hierarchies, and treat entity types as key prefixes.

Jul 6, 2026 7 min read

NATS JetStream Event-Driven Architecture Streams

NATS JetStream: When Events Need to Survive the Moment

JetStream adds durable streams and consumer delivery state so publishers and subscribers can be temporally decoupled — capture, replay, and redeliver until acknowledged.

Jul 6, 2026 9 min read

NATS Kafka AWS SQS Event-Driven Architecture

NATS vs Kafka vs AWS SQS: A Decision Framework for Engineers

An honest comparison of delivery semantics, persistence, ops burden, fan-out, replay, portability, and cost — with symptom-based selection tables instead of benchmark wars.

Jul 6, 2026 16 min read

Recent Articles

Topics

Browse by topic

Start with what you are building.

Start with what you are building.

Start with what you are building.

Start with what you are building.

Start with what you are building.

Start with what you are building.

Start with what you are building.

Start with what you are building.

Start with what you are building.

Start with what you are building.

Trends this month

Browse everything

Extend GitHub Spec Kit: Add Explore and Alternatives-Review Modes

How to Make GitHub Spec Kit Produce Leaner Output

Why GitHub Spec Kit Is So Interesting: Six Skills Worth Studying

Do New Frontier LLMs Really Resolve Ambiguity Better?

Let's Build Prompt-First Repositories for Your Applications

Claude Low Effort Compared: Fable 5 vs Opus 4.8 vs Sonnet 5

How to Implement and Evaluate a Low-Effort Model Routing Policy

Build an Ultracode-Like Coding Workflow with Agno and AgentOS

Build an Ultracode-Like Coding Workflow with LangGraph

DuckDB in Dev, Snowflake in Prod: A Practical dbt Workflow for a Medallion Project

Fable 5 High vs Opus 4.8 High vs Sonnet 5 High: A Practical Claude Model Guide

Claude Code Ultracode Architecture Blueprint for Cursor and Custom Agents

OpenSpec vs Superpowers: Spec Generation and Implementation Compared

WebAssembly in 2026: What It Means for AI, Vibe Coding, and the Future of the Internet

Agentic Workspaces: Submodules, Skills, Knowledge, and Spec-Driven Delivery

Running Agentic Workspaces in the Cloud at Scale

Session Retrospectives: Turning Agent Runs into Platform Improvements

Submodules as a Dependency Contract: Freshness Without Breakage

Trace, Triage, Debug: MCP Tooling for Agentic Workspaces

Pi (pi.dev): The Minimal Coding Agent Harness and How It Compares

Advanced Vibe Coding: A Modern Agentic Workflow for Beginners

Core NATS: Subjects, Pub/Sub, Request/Reply, and Queue Groups

EDA with NATS: One Truth, Many Read Models — CQRS Projections

EDA with NATS: Observability and SLOs for JetStream Projection Pipelines

JetStream KV Is a Change Stream, Not a Table

NATS JetStream: When Events Need to Survive the Moment

NATS vs Kafka vs AWS SQS: A Decision Framework for Engineers