skills.rest
sohamda
sohamdaCOMMUNITY
Explore all skills by this author →
Website𝕏@iamsohamGitHub

golden-principles

Defines ten operating principles governing agent behavior in repository-driven infrastructure workflows.

Updated May 12, 2026
Search Tags:#code review#azure#progressive disclosure#repository context#invariants#agent governance#operating principles
One-click install
npx skills add https://github.com/sohamda/apex-try-out-demo-repo --skill golden-principles-sohamda
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: golden-principles
Source: https://github.com/sohamda/apex-try-out-demo-repo/tree/main/.github/skills/golden-principles
Command: npx skills add https://github.com/sohamda/apex-try-out-demo-repo --skill golden-principles-sohamda

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? AI agents working in a repository need consistent operating rules so their outputs remain governed, reviewable, and reconstructible from committed files rather than scattered chat history or ad-hoc decisions. ## Core Features & Use Cases - Ten Agent Operating Principles: Covers repository-as-system-of-record, invariant enforcement, AVM-first Azure modules, security baselines, context scarcity, and mechanical rule enforcement. - Decision Framework: Provides per-principle compliance tests so agents and reviewers can check design choices against non-negotiable rules. - Progressive Loading: Ships digest and minimal variants plus a detailed references file so agents load only the depth they need. - Use Case: When designing a new agent or skill for Azure infrastructure work, read these principles first to decide whether a rule belongs in documentation, a linter, or a skill update. ## Quick Start Ask the agent to read the golden principles skill and evaluate whether a proposed instruction change complies with the ten operating principles.

Frequently Asked Questions about golden-principles

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What are the golden principles for AI agents in a repository?▼

They are ten operating rules covering repository-as-system-of-record, map-not-manual documentation, invariant enforcement, boundary validation, AVM-first Azure modules, golden paths, encoded feedback, context scarcity, progressive disclosure, and mechanical enforcement of rules.

How do I apply the golden principles when creating a new agent?▼

Read all ten principles first, identify which two or three dominate each design decision, run the per-principle compliance test, and change the implementation wherever it conflicts with a principle. Document any deliberate deviation in an ADR.

When should a rule become a linter check instead of documentation?▼

Principle 10 states that any rule expressible as a linter check, CI validation, or pre-commit hook should be automated rather than documented. Machines enforce rules more reliably than paragraphs of prose.

What is the difference between SKILL.md and references/principles.md?▼

SKILL.md holds a one-line summary of each principle for quick context loading, while references/principles.md contains the full rationale and per-principle tests. When the two disagree, the references file is canonical.

When should I not use the golden principles skill?▼

Do not use it for Azure infrastructure code generation, troubleshooting, or diagram creation. It governs agent behavior and operating philosophy only; task-specific skills handle execution work.

Related Skills

Explore More Software Engineering Skills→
bmad-code-orgbmad-code-org
52.5k

bmad-project-context

Creates and maintains verified AGENTS.md instruction blocks for AI coding agents in a repository.

Official
Advanced
microsoftmicrosoft
3.0k

skill-creator

Guides creation of effective skills for AI coding agents working with Azure SDKs and Microsoft Foundry.

Official
Advanced
Avihai-HAvihai-H

azure-defaults

Automate Azure infrastructure defaults for regions, tags, naming, and security.

Community
Advanced
emilianoeloiemilianoeloi
6

dev-harness-architect

Audits, diagnoses, and designs development harnesses for coding agents across any stack.

Community
Advanced