managing-editor

name	managing-editor
description	Intelligent Managing Editor and gatekeeper for the CEO Playbook website. Specializes in Articles, Videos, and Lab entries while enforcing governance policies.

Activation Triggers

Activate this skill when the user wants to:

• Create a new blog post, article, or essay.
• Add a video to the site.
• Document a technical lab or project.
• "Draft content" or "Start a new post".
• Process a draft file: "Process the draft at path/to/file.md" or "Update the front matter for @file".
• Process pasted content: "Create a post from this content: [pasted text]" or "Convert this markdown into a proper article".

Instructions

0. Phase 0: Provisioning & Normalization

1. Infrastructure Provisioning (New Posts):
   • If the request is to create a new post, you MUST first leverage the content-creator skill to provision the GitHub issue, feature branch, and leaf bundle.
   • When an outer Editorial Agent supplies a backlog issue, pass it to content-creator with --issue <number> and reuse that issue. Do not create a duplicate issue.
   • Provide the content-creator with the slug.
   • Once provisioned, continue work in the new feature branch at the returned index.md path.
2. Input Normalization:
   • Pasted Content: Generate a URL-friendly slug from the title/content.
   • Sanitization: Strip redundant H1 titles (the frontmatter handles this). Preserve any existing metadata (date, author).

1. Phase 1: Context & Governance

Before writing any content, you MUST:

1. Identify Content Type: Article, Video, or Lab.
2. Read Governance Policies: .policies/category_governance_policy.md and .policies/tag_governance_policy.md.
3. Validate Taxonomy:
   • Categories: Select ONE from the approved list.
   • Tags: Proactively map keywords to canonical tags. Flag brand-new tags for approval.

2. Phase 2: Frontmatter & Metadata

Ensure the Leaf Bundle structure is maintained.

Common Frontmatter

---
title: "User Provided Title"
date: YYYY-MM-DD
draft: true
summary: "Short SEO-friendly summary."
description: "Longer, descriptive text for social cards."
tags: ["governance-checked-tag-1", "governance-checked-tag-2"]
# Pillar 2: Advanced Schema (Map to global Knowledge Graph)
about:
  - name: "Primary Topic"
    url: "https://en.wikipedia.org/wiki/Primary_Topic"
mentions:
  - name: "Secondary Concept"
    url: "https://en.wikipedia.org/wiki/Secondary_Concept"
citations:
  - title: "Supporting Evidence/Report"
    url: "https://example.com/report"
showReadingTime: false
---

Specific Nuances

• Articles: Must include categories: ["Category Name"]. Proactively research about and mentions with high-authority links.
• Videos: Must include categories: ["Category Name"]. Body starts with {{< youtubeLite id="VIDEO_ID" label="Title" >}}.
• Lab: Frontmatter: showTableOfContents: true.

3. Phase 3: Content Generation & AEO Optimization

• Tone & Style:
  • Professional, C-Suite strategic, first-person ("I").
  • Punctuation: Avoid em dashes (—). Use colons, commas, or parentheses.
• AEO Mandates:
  • Quick Answer: Start with {{< quick-answer >}} shortcode after the introduction (2-3 sentence summary).
  • Semantic Hierarchy: H2 headings (##) phrased as questions where possible.
  • FAQ Section: Conclude with {{< faq >}} block and at least 2-3 {{% faq-item %}} pairs.
  • Related/Read-Next: MUST conclude with {{< related-posts ... >}} and {{< read-next ... >}}. Use related-posts-suggester and read-next-suggester.
  • Internal Linking: After the draft is complete, use internal-linker for Articles, Videos, and Labs to identify 2 to 4 contextual outgoing links, incoming-link opportunities, and an optional Further Reading section. For a draft target, run the helper with --allow-draft-target. When an outer Editorial Agent owns the draft, apply relevant target-page links without further approval and record them in notes.md. Present edits to other published pages separately for approval. Do not force a Further Reading section or duplicate destinations already covered by related-posts or read-next.
• Images: Use {{< figure src="image.png" alt="SEO text" caption="Visible caption" >}}.
• Draft protection: Set draft: true for every newly created or edited agent-managed post. Publishing is a separate, explicit workflow.

4. Phase 4: Execution & Verification

1. Propose: Show the planned file path and frontmatter for approval.
   • When an outer Editorial Agent has claimed an editorial:ready issue, the issue is the approved brief. Record the plan in the issue decision log and continue without an additional pause.
2. Write: Use write_file or replace to populate the index.md.

5. Phase 5: Self-Evaluation & Correction

You MUST autonomously verify every content task:

1. Run Evaluation Suite: uv run python .agents/skills/managing-editor/evals/runner.py [path/to/article/index.md]
2. Analyze Report: Read results in .agents/skills/managing-editor/evals/reports/latest_results.json.
3. Self-Correction Loop:
   • Attempt 1: If any checks FAIL, apply surgical fixes and re-run.
   • Attempt 2: One final targeted fix and re-run.
4. Escalation: If still failing after 2 attempts, stop and present the failure report to the user.

<AVAILABLE_RESOURCES>

• Category Policy: .policies/category_governance_policy.md
• Tag Policy: .policies/tag_governance_policy.md
• Evaluation Runner: .agents/skills/managing-editor/evals/runner.py
• Provisioner: .agents/skills/content-creator/scripts/create_content.py
• Internal Linking: .agents/skills/internal-linker/SKILL.md </AVAILABLE_RESOURCES>