DADL: A Declarative Description Language for Enterprise Tool Libraries in LLM Agent Systems

The Model Context Protocol (MCP) has rapidly established itself as the standard interface between large language model (LLM) agents and external tools. Adopting MCP at organizational scale, however, exposes two structural problems. First, every API integration is shipped as a dedicated server process with its own deployment, dependency tree, and credential handling, and recent empirical work shows that the overwhelming majority of these servers are thin wrappers around REST APIs. Second, the per-tool registration model causes context window consumption to grow linearly with the number of available tools, which forces real deployments to expose only a small fraction of the APIs an organization actually uses. We present DADL (Dunkel API Description Language), a YAML format that describes a REST API's endpoints, authentication, pagination, response shaping, and access classification in a single declarative file. A DADL file is interpreted by an execution layer at runtime; no per-API server process is deployed and no integration code is generated, although the shared runtime is itself a server. Because all tools share that one runtime, credentials are managed centrally, authorization decisions are made centrally, and the available tools are exposed to the LLM through a fixed-size Code Mode interface that is independent of catalog size. The combination produces what we call an Enterprise Tool Library: a versioned, auditable collection of API integrations that any team in an organization can extend, share, and consume through a common authentication and authorization boundary. The DADL v0.1 specification is released under CC BY-SA 4.0, and a public registry currently contains 1,833 tool definitions across 20 services. On this catalog, Code Mode reduces the LLM context cost of tool advertisement — the schemas the model must see in order to know what is available — from approximately 142,000 tokens to approximately 1,000, a 142x reduction; the per-call cost of search and execute invocations is additional and depends on the task.