About

About Us

About the Agent Ecosystem Research Program

Independent research and standards work for the agent ecosystem, led by a researcher with deep expertise in documentation infrastructure and developer tooling.

Dachary Carey

Lead Researcher

The agent ecosystem research grew from direct experience: watching agents fail while trying to retrieve information on hundreds of documentation sites, and realizing that nobody was systematically studying these failure modes or proposing solutions.

Kay Rhodes

Developer

Lending 20+ years of development experience to build the tooling registry, installation CLI, and other planned developer tools for the agent ecosystem.

Rhyannon Rodriguez

Poker-of-Things

Bringing a background combining library science, development expertise, and technical writing to ask questions, run tests, and document the results.

Our Principles

What Drives Us

Research that’s independent, practical, and open by default.

Independence

Sponsors support the program; they don’t direct conclusions. Published findings reflect what the data shows, not what sponsors prefer.

Empirical Rigor

Every specification and recommendation is built from direct observation and systematic evaluation, not assumptions.

Open by Default

Specifications, tools, and research methodologies are open source. Ecosystem standards should be available to everyone.

Open Source Projects

6

Agent Skills Audited

673+

Published Articles

13

FAQ

Frequently Asked Questions

Common questions about the Agent Ecosystem Research Program.

It’s an independent research program studying how coding agents interact with documentation, tools, and infrastructure. We publish findings, build open-source tooling, and develop specifications (like the Agent-Friendly Documentation Spec) that help the ecosystem mature. The program is already operational with published outputs and automated research infrastructure.

The research program has been bootstrapped to date, self-funded by the lead researcher and individual members. All published outputs, tooling, and infrastructure were built without external funding. We’re now seeking corporate sponsorship, grants, and community support to sustain and expand the work. Sponsors support the program but don’t direct conclusions. Published findings reflect what the data shows, not what sponsors prefer. That independence is what makes the research credible and useful.

It means documentation that coding agents (like Claude Code, Cursor, GitHub Copilot) can effectively discover, access, and use. Our 22-check specification covers areas like llms.txt availability, markdown content, page size limits, content structure, URL stability, and discoverability. When documentation meets these checks, agents can use it reliably; when it doesn’t, agents fail in predictable and preventable ways.

We’re setting up self-contained, time-bounded tasks suitable for volunteers: auditing documentation sites with our afdocs tool, researching MCP servers for upcoming reports, or co-authoring sections of published research. All contributions receive visible attribution. Check the Get Involved page for current opportunities.

Everyone building or using agent tooling. Agent platform companies benefit when documentation works well with their agents. Developer tool companies benefit when they understand how agents consume their docs. IDE and coding tool companies benefit from better tool quality and interoperability standards. And developers benefit from an ecosystem that just works.

Yes. Our specifications, audit tools (afdocs, skill-validator), and research methodologies are all open source. Published research and reports are freely available. We believe ecosystem standards should be open by default.

Help Build the Agent Ecosystem’s Infrastructure

This research is independent by design. Sponsors support the program; they don’t direct conclusions. That independence is what makes the research credible and useful to the industry.

Become a Sponsor