Programmatic SEO Skill Packet

Agent-ready skill packet for planning and executing programmatic SEO systems.

This packet is built for schema-first pSEO work: keyword patterns, taxonomy, data sources, JSON schemas, renderers, QA, progressive rollout, and monitoring. It is not for thin city-name or variable-swap pages.

Load Order

1. SKILL.md - primary agent instructions and output contract.
2. references/playbooks.md - page-family patterns and when to use each.
3. references/pseo-2.0-process.md - deeper architecture, rollout, and guardrails.
4. references/work-order-template.md - reusable work order for execution planning.
5. agents/openai.yaml - short OpenAI/Codex-facing descriptor.

Install

Clone this repository into an agent skill directory:

git clone https://github.com/michaelmcker/programmatic-seo.git ~/.agents/skills/programmatic-seo

Or point the agent directly at the repository and instruct it:

Use the Programmatic SEO skill packet. Load SKILL.md first, then references/work-order-template.md if this needs to become an execution plan.

Use When

• Programmatic SEO or pSEO strategy.
• Template pages.
• Pages at scale.
• Directory pages.
• Location or service-area pages.
• Comparison, integration, profile, product-variant, or entity page families.
• Repeated long-tail SEO opportunities where a schema and renderer can support many useful pages.

Do Not Use When

• The task is a one-off landing page, homepage, blog post, or normal service page.
• There are fewer than 10-20 viable pages.
• The idea only swaps variables such as city names.
• There is no differentiated data source.
• Nobody can monitor indexation, prune weak pages, or maintain the taxonomy.

Required Skill Output

When the skill runs, it should return:

1. Fit decision: proceed, narrow, or do not proceed.
2. Keyword pattern and intent summary.
3. Taxonomy dimensions and candidate page count.
4. Data source inventory and defensibility rating.
5. Page-family architecture.
6. JSON schema outline.
7. Renderer or template outline.
8. Internal-linking plan.
9. Indexation and rollout plan.
10. QA checklist.
11. Open risks and missing inputs.

Core Rule

AI content should be built, not written. The model fills strict JSON schemas with source-backed, variation-specific data. Renderers turn that structured data into pages.