Enforce claude md requirement

[ikrispin]

Summary

• Added validate_claude_md() to scripts/validate_structure.py - checks CLAUDE.md existence, required sections (Skill-First Rule, Intent Routing, MCP Servers, Global Rules), and intent routing completeness for every skill
• Updated root CLAUDE.md pack architecture tree to include CLAUDE.md as a standard file, and added it as a required step in "Creating a New Agentic Pack" and "Adding a Skill" workflows
• Added Design Principle 11 (Pack-Level CLAUDE.md) to SKILL_DESIGN_PRINCIPLES.md with required sections and update-on-new-skill rule

Pack(s) affected

• rh-sre
• rh-developer
• ocp-admin
• rh-virt
• rh-ai-engineer
• Other / repo-wide

Change type

• New skill
• New agent
• New pack
• Update existing skill / agent
• MCP server config (.mcp.json)
• Docs / README
• CI / tooling

CLAUDE.md compliance

• Agents orchestrate skills; no direct MCP/tool calls in agents
• Skills are single-purpose task executors
• Skills encapsulate all tool access (MCP tools invoked only inside skills)
• Document consultation: file is read with the Read tool, then declared to the user
• No credentials hardcoded; env vars used via ${VAR} references
• Human-in-the-loop confirmation added for any destructive or critical operations

Validation

• make validate passes locally
• New/changed skills have valid YAML frontmatter (name, description)
• New/changed agents have valid YAML frontmatter (name, description)

[r2dedios]

Good job! Just a couple of comments to check from my side :)

[r2dedios]

Does it make sense to include here also the model and color fields for the skill description?

[ikrispin]

Good point: cff4a26

[r2dedios]

I think this part is already covered by scripts/run-skill-linter.sh

[r2dedios]

Additionally, please review the actions feedback: https://github.com/RHEcosystemAppEng/agentic-collections/actions/runs/23442448018/job/68196883970?pr=48