Skip to content

Conversation

@ayogun
Copy link

@ayogun ayogun commented Jan 20, 2026

Pull Request

Description

Type of change:

  • New content (adding best practices, patterns, or guidance)
  • Fix (correcting outdated or incorrect information)
  • Refactor (reorganizing or improving clarity)
  • Documentation (README, CONTRIBUTING, etc.)
  • Testing framework improvement

Summary:

Testing Evidence (REQUIRED)

Scenarios Tested

  • Scenario #: [Name]
  • Scenario #: [Name]

Baseline Behavior (WITHOUT changes)

Prompt: [test prompt]

Agent response: [verbatim or screenshot]

Issues:
- [What was missing or incorrect]

Compliance Behavior (WITH changes)

Prompt: [same test prompt]

Agent response: [verbatim or screenshot]

Improvements:
- [What improved]
- [Patterns now followed]

Evidence of Improvement

  • Agent references new content
  • Agent applies new patterns proactively
  • Agent doesn't rationalize skipping guidance
  • No new rationalizations introduced

Standards Compliance Checklist

Frontmatter (if SKILL.md changed)

  • Only name and description fields present
  • Description starts with "Use when..."
  • Description focuses on triggers/symptoms (not workflow summary)
  • Description < 1024 characters
  • Total frontmatter < 1024 characters
  • Name uses only letters, numbers, hyphens

Token Efficiency

  • SKILL.md remains <1,500 words (current: ~1,400)
  • Detailed content moved to skills/*.md reference files where appropriate
  • Used tables instead of prose
  • No content duplication

Content Quality

  • Imperative voice ("Use X", not "You should use X")
  • Scannable format (tables, bullets, clear headers)
  • Code examples are complete and runnable
  • Version-specific features clearly marked
  • ✅ DO vs ❌ DON'T patterns where appropriate

File Organization

  • Core content in SKILL.md
  • Detailed guides in skills/*.md
  • Testing updates in tests/*.md
  • No new files outside standard structure

Validation

  • Frontmatter validation passes
  • File size within guidelines
  • No broken internal links
  • Markdown lint clean
  • No TODO/FIXME comments (or tracked in issues)

Rationalizations

New rationalizations discovered:

  • None discovered
  • Documented in tests/rationalization-table.md
    • Rationalization: [verbatim agent excuse]
    • Counter added: [how SKILL.md now addresses it]

Related Issues

Closes #
Relates to #

Additional Context


For Maintainers

Review Checklist

  • Testing evidence is convincing (baseline → compliance improvement shown)
  • Standards compliance verified
  • Content is accurate and current
  • Token efficiency maintained
  • Quality standards met
  • No conflicts with existing content
  • CHANGELOG.md updated (if needed)

Merge Checklist

  • Squash commits with clear commit message
  • Update CHANGELOG.md if not done in PR
  • Consider if version bump needed
  • Tag if this completes a planned milestone

Copilot AI and others added 4 commits January 20, 2026 05:10
Co-authored-by: ayogun <62607936+ayogun@users.noreply.github.com>
Co-authored-by: ayogun <62607936+ayogun@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
- Remove invalid terraform state mv command for moving between state files
- Remove dangerous cleanup script with hardcoded LOCK_ID
- Fix invalid backend_config block for Terraform Cloud lock timeout
- Fix invalid string interpolation in backend configuration

Co-authored-by: ayogun <62607936+ayogun@users.noreply.github.com>
Copilot AI and others added 3 commits January 20, 2026 11:43
Co-authored-by: ayogun <62607936+ayogun@users.noreply.github.com>
Co-authored-by: ayogun <62607936+ayogun@users.noreply.github.com>
…ect Lock

Co-authored-by: ayogun <62607936+ayogun@users.noreply.github.com>
Copy link
Owner

@antonbabenko antonbabenko left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Awesome additions. The main concern I have is the level of detail in SKILL.md - it should be refactored. Important information from SKILL.md should go to references/.

- [Atlantis](https://www.runatlantis.io/) - Terraform automation with locking

### Best Practices Guides
- [terraform-best-practices.com - State Management](https://www.terraform-best-practices.com/state)
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
- [terraform-best-practices.com - State Management](https://www.terraform-best-practices.com/state)

There is no such page.


**Use terraform_remote_state for component dependencies:**

```hcl
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let's drop this code snippet since it is already well-known for LLMs.

}
```

### Essential State Commands
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let's not have Terraform CLI commands in the SKILL.md itself, but have them in the quick-reference.md . They are there already, so drop this section from SKILL.md


### State Locking

**How it works:**
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LLM already knows "HOW", we need to focus on concrete solutions/workarounds and decision-making processes. Let's drop section "### State Locking".

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants