blog: guiding your agents

[Pertempto]

No description provided.

[github-actions]

Code Review: PR #13

Title Check

The PR title "blog: structuring AGENTS.md" follows conventional commit style (type: message format). ✅

Content Review

Strengths:

• Well-structured blog post with clear sections covering best practices
• The three-tier system (Always Do, Ask First, Never Do) is a practical framework
• Good use of external resources (arxiv study, addyosmani.com reference)
• Includes practical advice about avoiding documentation rot
• Mentions Agent Skills as an alternative to bloating AGENTS.md

Minor Suggestions:

• Line 30: The link to addyosmani.com has a long URL fragment. Consider using a URL shortener or cleaner link format for readability.
• The post could benefit from a brief code example or concrete scenario demonstrating the three-tier system in practice.

Overall: The content is informative and aligns well with the project's apparent focus on effective agent collaboration patterns. No blockers.

---

Review by opencode

---

🔗 OpenCode Session

[bambam955]

This is really good! I find the insight about the relationship between AGENTS.md and agent skills to be really really helpful.

I recently saw a post where someone said (kind of joking but kind of not) that "context engineering" would become the chief job of a software dev soon. I don't know if that's true or not, but this definitely gives me a good starting point for how to manage agents more effectively 💯 💯 💯