AI and LLM Resources
Alongside the human-readable documentation on this site, Netify publishes a parallel set of machine-readable resources for AI coding agents, LLM tooling, and OpenAPI-based codegen and API clients. This page indexes all of them for Netify Agent v5.
By providing structured, machine parsable context natively, these assets allow AI tools to bypass expensive site scraping or manual schema building, dramatically reducing hallucination risks and accelerating client generation. Developers can instantly ingest these specs into their pipelines to build, test, and deploy Netify v5 integrations in a fraction of the time.
LLM-Friendly Documentation
Netify follows the llms.txt convention: a curated, plain-text index designed to be read directly by an LLM, with short descriptions so an agent can decide what to fetch next without pulling in a full page of marketing HTML.
- llms.txt
- The curated index that provides configuration information, JSON schemas, telemetry and more.
- llms-full.txt
- Every page in llms.txt concatenated into a single file, for agents that prefer one fetch over many.
Each entry in llms.txt links to an individual Markdown document, JSON schema or OpenAPI specification. These are maintained specifically for machine consumption, and can differ from the corresponding human page's wording where that makes them clearer for an LLM.
OpenAPI Specification
The Integration API provides information on licensing, applications, signatures and more. The Integration API OpenAPI specification is an OpenAPI 3.1 document that covers this API in detail. Use it to generate typed API clients, feed an LLM tool-calling schema, or drive request validation.
JSON Schemas
Every processor and sink plugin's configuration file, and every telemetry record's payload, has a companion JSON Schema. The schema links are available in llms.txt, under Configuration Schemas (plugin configuration) and Telemetry Schemas (telemetry payloads).
Using These Resources
Point an AI coding agent or LLM tool at the llms.txt index and let it fetch specific docs or schemas as needed. Some tools accept a single system-prompt-level URL for project context; others expect files fetched on demand as a conversation progresses. Both work here - llms.txt is small enough to load up front, while llms-full.txt or a single schema file are better fetched only when needed. Prompt your AI tool on how to best use these files for the specific tasks.