Skip to content

chore: Introduce AI agent friendly skills - #522

Merged
mlakov merged 5 commits into
mainfrom
chore/introduce-ai-agent-friendly-skills
Jul 28, 2026
Merged

chore: Introduce AI agent friendly skills#522
mlakov merged 5 commits into
mainfrom
chore/introduce-ai-agent-friendly-skills

Conversation

@mlakov

@mlakov mlakov commented Jul 25, 2026

Copy link
Copy Markdown
Contributor

Introduce AI Agent-Friendly Skills for @cap-js/ord

New Feature

✨ Adds two AI agent skill definitions (compatible with tools like Claude's slash commands) to guide developers through integrating and configuring the @cap-js/ord plugin using natural language prompts.

Changes

  • .skills/integrate-ord-plugin/SKILL.md: New skill that walks AI agents through the full integration of @cap-js/ord into a CAP Node.js project. Covers:

    • Installation and auto-activation behavior
    • .cdsrc.json configuration grouped by purpose (Identity, Document metadata, Product/package ownership, Resource grouping, External dependencies, Advanced options)
    • Authentication setup (Basic auth, CF mTLS for production and local dev)
    • Verification steps
    • Supports argument-based dispatch (config, auth) for targeted help
  • .skills/custom-ord-annotations/SKILL.md: New skill documenting all CDS annotations used to customize the ORD document produced by @cap-js/ord. Covers:

    • Visibility control and propagation rules
    • Title and description resolution chains per resource type
    • ORD ID and version overrides
    • Package and consumption bundle assignment
    • Entity type configuration and @EntityRelationship.entityType gating
    • Data products (local primary and external integration dependencies)
    • OpenAPI server injection
    • Pass-through fields via @ORD.Extensions with spread behavior differences between API resources vs. event resources/entity types

Both skills are marked user-invocable: true, define allowed-tools to constrain AI agent actions, and use argument-based dispatch tables for targeted responses.

  • 🔄 Regenerate and Update Summary
PR Bot Information

Version: 1.29.2

  • Output Template: Default Template
  • File Content Strategy: Full file content
  • Correlation ID: c726d2c0-8844-11f1-9ba5-c07e43fbd797
  • Summary Prompt: Default Prompt
  • LLM: anthropic--claude-4.6-sonnet
  • Event Trigger: pull_request.edited

@mlakov
mlakov requested review from swennemers and zongqichen July 25, 2026 16:12
hyperspace-pr-bot[bot]

This comment was marked as outdated.

@mlakov mlakov added run-e2e Trigger e2e test pipeline ready for review labels Jul 25, 2026
@mlakov
mlakov force-pushed the chore/introduce-ai-agent-friendly-skills branch from 3c85ee2 to ef15eb2 Compare July 25, 2026 16:19
Comment thread .skills/custom-ord-annotations/SKILL.md Outdated
Comment thread .skills/custom-ord-annotations/SKILL.md Outdated
Comment thread .skills/integrate-ord-plugin/SKILL.md
Comment thread .skills/integrate-ord-plugin/SKILL.md
Comment thread .skills/integrate-ord-plugin/SKILL.md
Comment thread .skills/integrate-ord-plugin/SKILL.md Outdated
@mlakov
mlakov force-pushed the chore/introduce-ai-agent-friendly-skills branch from ae89274 to 14b659c Compare July 28, 2026 06:26
@mlakov

mlakov commented Jul 28, 2026

Copy link
Copy Markdown
Contributor Author

@swennemers - big thank you for the detailed review of the skills. I've addressed all your comments. If you have some time - can you, please, take a second look and maybe also an approve ?

@mlakov
mlakov enabled auto-merge (squash) July 28, 2026 08:06
@mlakov
mlakov force-pushed the chore/introduce-ai-agent-friendly-skills branch from 47c5ef4 to 572e893 Compare July 28, 2026 08:09
@mlakov
mlakov merged commit 7269a1d into main Jul 28, 2026
5 checks passed
@mlakov
mlakov deleted the chore/introduce-ai-agent-friendly-skills branch July 28, 2026 08:49
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ready for review run-e2e Trigger e2e test pipeline

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants