Beacon / Machine Outputs
Machine-Readable Outputs
Everything on this site is generated from a single structured graph. These are the machine-readable surfaces that AI agents, search engines, and automation tools can consume directly.
The Beacon Thesis
One graph in. Many surfaces out.
Most companies maintain their public presence as a sprawl of disconnected artifacts that never stay in sync. The human site says one thing, the agent surfaces say another, and the search index reflects something else entirely.
Beacon solves this by holding one canonical, structured representation of everything — then emitting every artifact each reader needs from that single source. Humans see audience-tailored views. Agents see machine-readable surfaces. Everything regenerates from one edit. Drift goes to zero.
Available Outputs
llms.txt
Plain TextA plain-text file that describes Sightbox in a format optimized for LLM consumption. Contains company overview, offerings, pricing, and contact information.
/llms.txt
JSON-LD Schema
JSON-LDSchema.org structured data for Organization, Service, and FAQPage. Embedded site-wide for search engines and knowledge graphs.
Structured Data
Narrative Layer
Beacon+A bespoke, authored narrative — story, positioning, and voice tuned for the agent reader. Narrative architecture applied to the agent layer; the premium, uniquely-Sightbox tier that sits on top of the generated surfaces.
Authored · premium
MCP Server
Coming SoonModel Context Protocol server for AI agents to query Sightbox data programmatically. Enables tools like Claude, ChatGPT, and custom agents to retrieve structured information.
/.well-known/mcp
Full Output Catalog
Outputs mapped to agent-readiness standards (Fern, Mintlify, IsItAgentReady):