AgentFriendly for AI Agents
What AgentFriendly is
AgentFriendly is a website scanner that evaluates how reliably AI agents can discover, parse, and recommend public web content.
What problem it solves
Many websites are readable for humans but fragile for non-rendering bots and retrieval systems. AgentFriendly surfaces concrete issues that degrade machine reliability.
Who it is for
Teams that want their product and docs to be accurately represented by AI agents.
Who it is not for
It is not a rank tracker, backlink platform, or ad optimization suite.
How it works
It crawls up to a capped number of pages, inspects raw HTML and metadata, scores by category, and returns prioritized findings plus remediation guidance.
Limitations
Results are heuristic, no LLM speculation is required, and dynamic behavior is only partially analyzed unless Playwright checks are enabled.
Pricing / usage model
Open MVP for local use. No hosted billing in this repository.