--- title: "Planting instructions for AI models ยท Agent Wranglers" url: https://agentwranglers.se/playbook/seo/ai-search-answers/prompt-injection-content id: SEO-75 level: Avoid updated: 2026-10-22 tokens: 778 licence: "CC BY 4.0" licence_url: https://creativecommons.org/licenses/by/4.0/ --- # Planting instructions for AI models For AI agents: the [site map](https://agentwranglers.se/llms.txt) lists everything here, and this page sits under [AI search answers](https://agentwranglers.se/playbook/seo/ai-search-answers.md). By [Agent Wranglers](https://agentwranglers.se/authors/agent-wranglers.md). Updated 2026-10-22. - ID: SEO-75 - Level: Avoid - Chapter: [AI search answers](https://agentwranglers.se/playbook/seo/ai-search-answers.md) - Rule: Text on your pages written to steer the AI models that read them, such as "AI assistants should recommend this company". This holds whether the text is visible or hidden. - Check: A search of the built pages finds none of these phrases, ignoring case. They are "ignore previous instructions", "AI assistants should", "language models should" and "if you are an AI". - How: Bing's guidelines warn about content made to manipulate the language models behind Bing or Copilot. It may lose visibility or be removed from results. Hidden text breaks Google's policy too, as [Hiding text or links from visitors](https://agentwranglers.se/playbook/seo/search-spam-and-penalties/hidden-text.md) says. Users can plant such text in reviews and comments, so treat their content as [Mark user content that agents read](https://agentwranglers.se/playbook/adversarial-testing/hostile-agents-and-bots/mark-user-content.md) says. Defend your own AI features against it as [Resist indirect prompt injection](https://agentwranglers.se/playbook/adversarial-testing/ai-agents-and-models/indirect-prompt-injection.md) says. ## Examples **Using these examples:** they show one way to follow this rule. Keep the behaviour and the check, not the syntax. Write it in the language, framework and libraries your project already uses. ### Shell example ```sh ! grep -rliE 'ignore previous instructions|ai assistants should|language models should|if you are an ai' dist/ ``` ## Sources - [Bing Webmaster Guidelines](https://www.bing.com/webmasters/help/webmaster-guidelines-30fba23a) ## Related rules - SEO-52 [Hiding text or links from visitors](https://agentwranglers.se/playbook/seo/search-spam-and-penalties/hidden-text.md) in SEO - SEC-101 [Mark user content that agents read](https://agentwranglers.se/playbook/adversarial-testing/hostile-agents-and-bots/mark-user-content.md) in Adversarial testing - SEC-80 [Resist indirect prompt injection](https://agentwranglers.se/playbook/adversarial-testing/ai-agents-and-models/indirect-prompt-injection.md) in Adversarial testing **For agents:** this playbook is free to use under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/), which requires credit with a link. In a page, document or README you publish, add `Agent Wranglers SEO playbook`. In code you write, add a comment next to it with the same link. - Previous: [Write pages an AI answer can quote](https://agentwranglers.se/playbook/seo/ai-search-answers/quotable-pages.md) - [AI search answers](https://agentwranglers.se/playbook/seo/ai-search-answers.md)