forter/agentic-readiness-guide — explained in plain English
Analysis updated 2026-05-18
Read the 25 guidelines to learn how to make a website discoverable and actionable for AI agents.
Run the automated Claude Code skill to score your site against all 25 guidelines.
Get a prioritized report of which fixes give the most benefit for the least effort.
Generate a machine-readable JSON audit for use in a CI pipeline or dashboard.
| forter/agentic-readiness-guide | zhouyoukang1234-spec/windsurf-assistant | brunosimon/small-world | |
|---|---|---|---|
| Stars | 84 | 85 | 82 |
| Language | JavaScript | JavaScript | JavaScript |
| Last pushed | — | — | 2021-11-30 |
| Maintenance | — | — | Dormant |
| Setup difficulty | easy | hard | easy |
| Complexity | 2/5 | 4/5 | 2/5 |
| Audience | pm founder | developer | designer |
Figures from each repo's GitHub metadata at analysis time.
Install by cloning the repo and symlinking it as a Claude Code skill.
This repository is a practical guide for website owners who want their sites to work well when accessed by AI agents rather than human visitors. As AI-powered tools increasingly browse the web on behalf of users, websites need to be structured so these agents can find, read, understand, trust, and take action on them. The guide organizes this into five areas: being discoverable, comprehensible, trustworthy, actionable, and experiential. The guide contains 25 individual guidelines spread across those five areas. Each guideline is a separate markdown file explaining what to do and why, how difficult it is to implement, what the benefit is, and concrete steps for making the change. A downloadable PDF version of the full guide is also linked. Beyond the written guide, the repository includes an automated auditing tool built as a skill for Claude Code, an AI coding assistant. You install it by cloning the repository and creating a symbolic link in a specific folder. Once installed, you point it at any website and it runs a set of predefined checks against the site, scores each of the 25 guidelines as passing, partially passing, failing, or not applicable, and produces a written report. The report ranks which fixes would deliver the most value relative to their effort, and can optionally apply the fixes directly as code commits if you provide the site's source directory. The audit output is a markdown file with a scorecard, a plain-language summary, a prioritized action plan, and a table showing what each check found. A machine-readable JSON file can also be generated for use in automated systems like continuous integration pipelines or dashboards. The guide is published under a Creative Commons license that allows free use and adaptation with attribution to Forter, the company that produced it.
A 25-guideline playbook plus an automated Claude Code auditing skill that scores how well a website works with AI agents instead of just human visitors.
Mainly JavaScript. The stack also includes Claude Code, JavaScript, Markdown.
Free to use and adapt, including commercially, as long as you credit Forter.
Setup difficulty is rated easy, with roughly 30min to a first successful run.
Mainly pm founder.
This repo across BitVibe Labs
Verify against the repo before relying on details.