# agent-skills | Addy Osmani | Knowledge Base | Instagit

Production-grade engineering skills for AI coding agents.

GitHub Stars: 16.2k

Repository: https://github.com/addyosmani/agent-skills

---

## Articles

### [Source-Driven Development Approach: A Framework Decision Methodology for Production Code](/addyosmani/agent-skills/what-is-the-source-driven-development-approach-for-making-framework-decisions)

Discover the source-driven development approach a four-step methodology that uses official documentation for accurate, traceable, and future-proof framework decisions in your production code.

- Tags: how-to-guide
- Published: 2026-04-16

### [How Progressive Disclosure Minimizes Token Usage in Agent-Skills](/addyosmani/agent-skills/how-does-progressive-disclosure-minimize-token-usage)

Learn how progressive disclosure minimizes token usage by loading essential skill entry points first and deferring large files until needed. Optimize your agent workflows now.

- Tags: performance
- Published: 2026-04-16

### [Context Engineering for Agent Memory Management: A Complete Guide to the Agent-Skills Methodology](/addyosmani/agent-skills/what-is-context-engineering-for-agent-memory-management)

Master context engineering for AI agent memory management. Learn how to curate agent information effectively to boost output quality and stay within token limits with the Agent-Skills methodology.

- Tags: deep-dive
- Published: 2026-04-16

### [How Agent Personas Like the Code-Reviewer Operate in agent-skills](/addyosmani/agent-skills/how-do-agent-personas-like-the-code-reviewer-operate)

Learn how agent personas like the code-reviewer operate in agent-skills by combining static Markdown role definitions with reusable skill frameworks for deterministic code reviews.

- Tags: how-to-guide
- Published: 2026-04-16

### [Migration Pattern for Handling Deprecations in Agent-Skills: A Four-Step Guide](/addyosmani/agent-skills/what-is-the-migration-pattern-for-handling-deprecations)

Learn the addyosmani/agent-skills four-step migration pattern for handling deprecations. Build replacements, document, migrate incrementally, and safely remove old systems.

- Tags: migration-guide
- Published: 2026-04-16

### [Feature Flag Lifecycle Management in Shipping and Launch Processes: A Complete Guide](/addyosmani/agent-skills/how-does-feature-flag-lifecycle-management-work-in-shipping-and-launch-processes)

Master feature flag lifecycle management from spec to deployment in the OpenCode model. Ship code safely with instant rollbacks and improve your launch processes.

- Tags: how-to-guide
- Published: 2026-04-16

### [Quality Gates in CI/CD Automation: The Complete Pipeline Guide](/addyosmani/agent-skills/what-are-the-quality-gates-in-ci-cd-automation)

Discover quality gates in CI CD automation; essential checkpoints like linting testing and security audits that ensure only defect free code reaches production.

- Tags: how-to-guide
- Published: 2026-04-16

### [How Trunk-Based Development Functions in the Git-Workflow-and-Versioning Skill](/addyosmani/agent-skills/how-does-trunk-based-development-function-within-git-workflow-and-versioning)

Learn how trunk-based development works in git-workflow-and-versioning. Maintain a deployable main branch with short-lived feature branches to reduce merge risks and boost performance.

- Tags: how-to-guide
- Published: 2026-04-16

### [Hyrum's Law in API Design: Managing Accidental Dependencies and Observable Behaviors](/addyosmani/agent-skills/what-is-hyrums-law-and-its-implications-for-api-and-interface-design)

Understand Hyrum's Law in API design. Learn how to manage accidental dependencies and observable behaviors to build more robust and maintainable APIs.

- Tags: deep-dive
- Published: 2026-04-16

### [How the Three-Tier Boundary System (Always/Ask First/Never) Operates in Agent-Skills](/addyosmani/agent-skills/how-does-the-three-tier-boundary-system-operate)

Understand the three-tier boundary system in addyosmani/agent-skills Always Ask First Never. Learn how it enforces safe, spec-driven automation with auto-execution, human approval, or forbidden actions.

- Tags: deep-dive
- Published: 2026-04-16

### [Chesterton's Fence in Software Engineering: Safely Simplifying Legacy Code](/addyosmani/agent-skills/what-is-chesternons-fence-and-its-application-in-code-simplification)

Understand Chesterton's Fence in software engineering. Learn how this principle helps safely simplify legacy code by avoiding accidental regressions during refactoring. Protect your codebase.

- Tags: best-practices
- Published: 2026-04-16

### [How the Five Axes of Code Review Function in the code-review-and-quality Skill](/addyosmani/agent-skills/how-do-the-five-axes-of-code-review-function-in-code-review-and-quality)

Understand the five axes of code review Correctness Readability Architecture Security and Performance within the code-review-and-quality skill for better code quality assessments.

- Tags: deep-dive
- Published: 2026-04-16

### [How DAMP Applies Over DRY in Test Code: A Complete Guide](/addyosmani/agent-skills/how-does-damp-apply-over-dry-in-test-code)

Discover how DAMP prioritizes readable test stories over DRY code reuse. Understand test code without helper modules. Learn more in this complete guide.

- Tags: deep-dive
- Published: 2026-04-16

### [What Is the Beyoncé Rule in Testing? A Complete Guide](/addyosmani/agent-skills/what-is-the-beyonce-rule-in-the-context-of-testing)

Discover the Beyoncé Rule in testing Your code should have a test if you liked and shipped its functionality making automated testing a personal accountability moment.

- Tags: deep-dive
- Published: 2026-04-16

### [How the Test Pyramid Works with the 80/15/5 Ratio](/addyosmani/agent-skills/how-does-the-test-pyramid-work-with-its-80-15-5-ratio)

Understand the test pyramid's 80/15/5 ratio for fast, reliable, and low-maintenance automated testing. Optimize your test suite with this key strategy.

- Tags: deep-dive
- Published: 2026-04-16

### [What Is the Prove-It Pattern for Bug Fixes in TDD?](/addyosmani/agent-skills/what-is-the-prove-it-pattern-for-bug-fixes-in-tdd)

Learn the Prove-It pattern for bug fixes in TDD. Write a failing test, implement a minimal fix, and ensure defects become regression tests. Master TDD bug fixing.

- Tags: deep-dive
- Published: 2026-04-16

### [How the Test-Driven Development Skill Implements the Red-Green-Refactor Cycle](/addyosmani/agent-skills/how-does-the-test-driven-development-skill-implement-the-red-green-refactor-cycle)

Learn how the test-driven development skill implements the Red-Green-Refactor cycle with AI agents. Follow explicit steps for writing failing tests, implementing code, and refactoring.

- Tags: internals
- Published: 2026-04-16

### [Spec-Driven Development Workflow: A 4-Phase Gated Process for Agent-Skills](/addyosmani/agent-skills/what-is-the-workflow-for-the-spec-driven-development-skill)

Explore the spec-driven development workflow for agent-skills. Learn the four gated phases Specify Plan Tasks and Implement that require human approval for agent advancement.

- Tags: architecture
- Published: 2026-04-16

### [How Slash Commands Map to the Development Lifecycle in the Agent-Skills Repository](/addyosmani/agent-skills/what-is-the-relationship-between-slash-commands-and-the-development-lifecycle)

Discover how slash commands in the agent-skills repo streamline the development lifecycle, mapping directly to each phase from definition to deployment.

- Tags: how-to-guide
- Published: 2026-04-16

### [How the Anti-Rationalization System Prevents Skill Skipping in Agent Workflows](/addyosmani/agent-skills/explain-the-anti-rationalization-system-and-how-it-prevents-skill-skipping)

Discover how the anti-rationalization system in addyosmani/agent-skills prevents skill skipping. Learn how it ensures agents complete every task lifecycle step for robust workflows.

- Tags: internals
- Published: 2026-04-16

### [Intent to Skill Mapping in AGENTS.md: How AI Agents Route Requests to OpenCode Skills](/addyosmani/agent-skills/how-is-intent-to-skill-mapping-handled-within-agents-md)

Discover how AI agents map user intents to OpenCode skills. AGENTS.md uses a declarative table for seamless routing without slash commands.

- Tags: how-to-guide
- Published: 2026-04-16

### [SKILL.md File Structure and Format in Agent-Skills: Complete Guide](/addyosmani/agent-skills/what-is-the-expected-structure-and-format-for-skill-md-files)

Learn the SKILL.md file structure and format for Agent-Skills. Discover how to define skill name, description, overview, when to use, core process, and verification for automatic agent discovery.

- Tags: how-to-guide
- Published: 2026-04-16

### [How the OpenCode Skill Execution Model Functions in agent-skills](/addyosmani/agent-skills/how-does-the-opencode-skill-execution-model-in-agent-skills-function)

Understand the OpenCode skill execution model in agent-skills. It automatically discovers selects and executes skills by detecting user intent and invoking the skill tool.

- Tags: internals
- Published: 2026-04-16