Spec tools and hooks: what they automate

Follow an ordinary repair, a shared dependency change and stale evidence to see where specification tools help—and which decisions remain with the reviewer.

Keeping specifications in the change review

Follow ordinary repairs, shared-component changes and stale evidence without turning every maintenance task into a new specification project.

Writing specifications for coding agents

A compact structure for specifying settled behaviour, preserving implementation discretion and keeping unresolved decisions in the working record.

Spec-driven development: an invoice export walkthrough

Follow a bounded invoice-export investigation from hidden decisions through agent evidence and separate specification failures.

Reviewing agent work and correcting specifications

How to check that an agent can use a specification, review bounded implementation evidence, and reconcile faulty or changed requirements with active work.

BDD and acceptance evidence for coding agents

How BDD can expose disagreements, how acceptance evidence earns its authority, and when to implement, investigate or revise a specification.

Spec-driven development without waterfall

Choose discovery, investigation or a bounded implementation before writing the next contract for a coding agent.

Dependency Lifecycle Management in Go

Introduction

Lifecycle management ensures that application components like servers, databases, and background workers are correctly initialized, started in the right order, and gracefully stopped.

Go developers typically wire dependencies manually, instead of relying on heavy frameworks or reflection based dependency injection. This approach improves readability and reduces hidden behavior, but it also shifts the burden of managing component interactions and lifecycle sequencing onto the developer.

As applications scale, the lack of built-in orchestration increases the risk of subtle bugs and inconsistent shutdown behavior.

structured logging: slog

Event logging is one of the three pillars of observability (traces and metrics being the other two).

A log entry is a timestamped record about the application’s activity, which can be used for troubleshooting, monitoring or auditing purposes.

Logs may have different formats:

• unstructured or plain text: similar to print statements, free-form text
• semi-structured: some values have a structure and the remaining data free-form
• structured: there is a well-defined and consistent format, such as: W3C Log Format, NCSA Log Format, Key/Value Pair, CSV or JSON, among others. Structured logs can be easily machine parsed, which simplifies analytics, filtering and aggregation.

This article is a non-exhaustive walkthrough of logging in Go.

Opentelemetry - Log-Based Change Data Capture tracing

Opentelemetry provides vendor neutral standards and implementations to generate, collect, and export tracing data.

A Context is a propagation mechanism which carries execution-scoped values across API boundaries and between logically associated execution units. Cross-cutting concerns access their data in-process using the same shared Context object.

Context propagation is required when the tracing needs to cross process or service boundaries. Common ways to propagate the context is using W3C Trace Context or Zipkin B3 headers, while to inject the context is, for example, http headers or metadata fields in event messages.