Why Standardize Claude Now

Without a standard approach, AI outputs can vary greatly based on the person and the prompt. Skills (folders containing a SKILL.md) and CLAUDE.md provide Claude with consistent guidance: uniform behaviours for non-development tasks, and ongoing engineering conventions for code. This leads to fewer one-time prompts and more repeatable, reviewable results. Claude+1

What are Claude Skills (and SKILL.md)?

Agent Skills are modular abilities that Claude can automatically apply when your request matches a Skill’s description. Each Skill includes a SKILL.md with instructions and optional scripts/templates, and Claude selects and uses relevant Skills without the need for slash commands. Think of them as “reusable playbooks” instead of one-off prompts. Claude+1

Where They Help

• Drafting policies, creating briefs, managing research, QA checklists (for non-development teams)

• Repetitive founder/operations workflows (such as triage, outreach, and competitive assessments)

• Consistent UI/UX review criteria for product design

Anthropic’s public examples and documentation illustrate how Skills package expertise so Claude knows when to use them. Claude

What is CLAUDE.md (for Engineering Teams)?

The CLAUDE.md file is read automatically by Claude Code to understand your project’s structure, coding standards, testing rules, commit/style conventions, and workflows. You can place it at the repository root (recommended) or where you run claude, and additional CLAUDE.md files can be added within sub-directories for local rules. Anthropic

Why It Matters: It eliminates the need for “cold start” prompting in each session. Claude Code begins with your context and writes diffs/tests/docs that align with how your team operates. Anthropic’s guidance and product blog provide information on how to structure these files for maximum effectiveness. Anthropic+1

Where to Place Domain Knowledge (Projects)

For broader context (such as brand voice, policies, playbooks, and research), use Projects to equip Claude with files and references. Projects create self-contained workspaces with their own knowledge base and chat history—perfect for marketing, support, or client work. Anthropic+1

Prompting Foundations (Quick Wins)

From Anthropic’s prompting best practices:

• Clearly state the audience, objective, constraints, and success criteria.

• Provide examples (both good and bad) and data snippets for reference.

• Request verification steps and citations where applicable.

• Avoid “over-engineering” by explicitly asking for minimal changes when coding. Claude

SKILL.md: A Practical Template

Use this framework to stabilize outcomes for non-development tasks (policies, briefs, QA):

# SKILL.md — InfoSec Policy Drafting (v1.0)
## Purpose
Draft or update InfoSec policies aligned to ISO 27001; produce reviewer-ready outputs.

## Triggers (Claude decides when relevant)
- “Draft/update InfoSec policy …”
- “Create a procedure for …”
- “Map controls to ISO/SoA …”

## Inputs
- Policy topic, scope, audience
- Constraints (standards, legal refs)
- Source docs (URLs/files)

## Steps
1) Clarify scope (ask 3–5 questions if ambiguous).
2) Outline sections; confirm in 5 bullets.
3) Draft with headings, definitions, RACI.
4) Add ISO mappings + change log.
5) Produce summary and next actions.

## Outputs
- Policy (Markdown, headings, numbered sections)
- 200-word exec summary
- Review checklist

## Refusal & Escalation
- Refuse legal advice; flag to Legal/SME when regulatory claims are requested.

## Evaluation Checklist
- Factuality vs supplied sources ✓
- ISO mapping included ✓
- Reading age ≤ 11–12 ✓

Skills are model-invoked and based around SKILL.md along with optional scripts/templates. Claude+1

CLAUDE.md: Engineering Starter

Add this to your repository root and modify depending on your stack:

# CLAUDE.md — Web App Conventions (v1.0)
## Architecture
Monorepo: /apps/web (Next.js), /packages/ui (React), /packages/config.
Testing: Vitest + Playwright. State: Zustand. Styling: Tailwind.

## Coding Rules
- Prefer small diffs; no unrequested refactors.
- Add/modify tests for changed logic.
- Accessibility: label form controls; keyboard nav required.
- i18n: wrap user-visible strings.

## Git & Reviews
- Conventional commits; one issue per PR.
- Include: summary, rationale, screenshots, test notes.
- Don’t add dependencies without a clear reason.

## DX Commands
- `pnpm dev`, `pnpm test`, `pnpm e2e`

## Definition of Done
- Tests pass; coverage not lower.
- Lighthouse perf ≥ 90 on changed pages.

## Prompts for You (Claude)
- Before coding: propose plan in bullets; wait for “OK”.
- Keep solutions minimal and focused; avoid over-engineering

Official guidance notes that Claude Code automatically reads CLAUDE.md and supports placing it at the repository root or wherever you run claude. Anthropic

Design Patterns (Worked Examples)

• Front-end Review Skill: ensure accessibility, performance budgets, design tokens, and tone of copy; produce a pass/fail checklist and annotated diffs. (Implement as SKILL.md plus an optional audit script.) Claude

• Founder Ops Skill: triage incoming items, enrich with web citations, and generate outreach copy under the brand voice; produce a CSV for CRM import.

• Policy Desk Skill: transform SME notes and standards into versioned SOPs with change logs and RACI.

Roll-out Checklist (6 Weeks)

1. Select Use Cases (2 to 3 per team) and build initial SKILL.md/CLAUDE.md.

2. Set Guardrails (for PII, client secrets, regulatory boundaries; approved sources; citation rules).

3. Pilot with 5 to 10 users per team; enable Projects for shared context. Claude Help Center

4. Measure: time to first draft, review cycles, defect rate, percentage of outputs with acceptable citations.

5. Harden: refine Skills, expand CLAUDE.md, add evaluation checklists.

6. Standardize: document in your playbook, train, and publish across the organization.

Tip: When coding, explicitly instruct for minimal changes to avoid unnecessary abstractions. Claude

Governance & Risk

• Traceability: require sources and citations for content directed at external audiences.

• Change Control: version SKILL.md/CLAUDE.md; PR review by subject matter experts.

• Access: separate Projects by client or department; apply the principle of least privilege. Claude Help Center

  
  

FAQ

What’s the Difference Between Skills and Prompts?
Skills are reusable, documented behaviours packaged with SKILL.md and optional resources; Claude can invoke them when relevant. Prompts are one-off instructions. Claude

Where Should I Place CLAUDE.md?
At the repository root (recommended) or wherever you run claude; use additional CLAUDE.md files in sub-directories for local rules if needed. Anthropic

How Do We Centralize Knowledge for a Team or Client?
Create a Project and upload reference documents, style guides, and past work; keep Skills small and focused, and point them at the Project’s materials. Anthropic+1