OpenSEO

Link Prospecting Agent Skill

Find relevant link prospects from SERPs, competitor backlinks, and topical publishers with your AI agent.

Run this skill in your agent
>/link-prospecting

The Link Prospecting Agent Skill finds relevant outreach prospects for a page or asset.

Your agent searches resource pages, articles, directories, and competitor backlink patterns. It uses OpenSEO data and web research to find sites with a real reason to mention your asset.

You get a smaller set of qualified prospects, with notes on why each one fits and what outreach angle could make sense.

What this skill helps your agent do

  • Build prospecting queries from your asset, topic, audience, or competitors.
  • Find resource pages, listicles, comparisons, directories, publishers, and topical articles.
  • Use competitor backlink patterns to understand where attention comes from.
  • Qualify prospects by topical relevance and outreach angle.
  • Draft outreach messages tied to the specific page and reason.
  • Avoid inventing contacts or treating spammy prospects as useful.

When to use it

Use this skill when you already have something worth linking to: a useful guide, product page, original study, tool, template, dataset, or strong point of view.

It also helps after competitor analysis if backlink patterns show publications, directories, communities, or partners that matter in your market.

What you get back

Your agent should return the best outreach angle, highest-priority prospect type, a prospect table, contact-discovery notes, and a few reusable outreach drafts.

How to get the best result

  • Share the target URL or describe the linkable asset.
  • Explain why someone would reference it.
  • Provide competitors when competitor backlinks are useful.
  • Ask the agent to prioritize relevance over volume.
  • Require source URLs for any contact details it finds.

Use it with OpenSEO MCP

Set up OpenSEO MCP so your agent can inspect SERPs, backlink overview data, and domain context before researching contact paths.

Read the actual skill

The source SKILL.md lives on GitHub: