Updated: October 7, 2026 · Source-first reference130 documented agents & platforms · No paid rankings

About AgentenCode

Independent intelligence for AI agents.

AgentenCode helps people understand, compare and verify AI agents through primary sources, structured evidence and transparent change history.

What AgentenCode is

AgentenCode is a reference and intelligence layer for AI agents and agentic systems. Instead of ranking products by hype, the project documents what can be supported by primary sources, what remains unknown and what has changed over time.

What the platform contains

130agent profiles
1,221evidence claims
802trust signals
250history events

The public platform combines profile pages, trust and EU context, structured history, explicit relations, news based on verified changes and machine-readable datasets.

Editorial principles

primary sources have priority. Unknown is kept distinct from false. Commercial relationships do not change the evidence rules. Legal and compliance conclusions are not inferred from isolated technical controls.

German and English

The German and English pages are two presentation layers over the same source-first data core. Research, evidence and history are maintained once; language-specific titles, explanations and editorial copy are rendered separately.

Why the project exists

AI-agent products change quickly and are often described with broad marketing language. AgentenCode reduces that ambiguity by tying important product statements to source-backed fields and by keeping a public record of verified changes.

The goal is not to declare a universal winner. It is to make evaluation more traceable for users, developers, researchers and organizations that need to understand what is documented, what remains uncertain and which source supports a decision-relevant claim.

Independence and product scope

AgentenCode does not sell ranking positions and does not convert commercial popularity into a trust score. The platform focuses on documentation quality, source traceability and practical decision context. A product with fewer documented controls is not automatically worse; it may simply have a larger public evidence gap.

The project covers individual agents as well as agentic platforms and system-level relationships. As the ecosystem evolves, the goal is to keep identity, evidence, trust and history separate enough that new product categories can be added without rewriting older facts.

Who the platform is for

AgentenCode is designed for people who need more than a product list: individual users comparing tools, developers examining technical capabilities, researchers following the ecosystem, and organizations reviewing governance, security or procurement questions.