Agentic & Machine-to-Machine Discovery
Relevant source files
The following files were used as context for generating this wiki page:
Agentic and Machine-to-Machine (M2M) Discovery provides structured, machine-readable data to autonomous agents, Large Language Models (LLMs), and AI search engines. This layer ensures that AI crawlers accurately interpret the technical specifications, service parameters, and entity relationships of Sunrise Communication.
The discovery architecture utilizes standardized manifest files, granular JSON-LD schema graphs, and specific crawler permissions to facilitate Artificial Intelligence Optimization (AIO) and Business-to-Agent (B2A) interactions.
Sources: task.md:1-12, sunrise-angular/public/llms-full.txt:1-8
Discovery Architecture
The discovery layer consists of multiple entry points designed for different types of machine consumers. While standard search engines use sitemap.xml, AI agents utilize llms.txt and llms.json for high-density information extraction.
The diagram shows how AI agents navigate from initial crawling permissions to structured manifest files and semantic schema data.
Sources: task.md:8-12, sunrise-angular/public/robots.txt:58-62
Machine-Readable Manifests
Sunrise Communication implements the /llms.txt standard to provide a condensed, text-based overview of the business for LLM context windows. This is supplemented by a full technical manifest and a JSON schema.
Discovery Endpoints
| Endpoint | Format | Primary Actor | Description |
|---|---|---|---|
/llms.txt | Markdown | LLM Context | High-level summary of company services and expertise. |
/llms-full.txt | Markdown | AI Agents / B2B | In-depth technical specs, hardware matrices, and AMC terms. |
/llms.json | JSON | Machine Systems | Structured manifest following llmstxt.org schema for automated discovery. |
/robots.txt | Text | All Crawlers | Permissions for AI user-agents and pointers to discovery files. |
Sources: task.md:9-12, sunrise-angular/public/llms.json:1-5
LLM Manifest Configuration (llms.json)
The llms.json file defines the business entity, authorized partnerships, and technical service scopes in a format compatible with autonomous procurement systems.
{
"$schema": "https://llmstxt.org/schema.json",
"name": "Sunrise Communication",
"reputation": {
"googleRating": 5.0,
"verifiedReviews": 93
},
"sla": {
"breakdownResponseTimeHours": 24,
"emergencySupportHours": "24/7"
}
}
Sources: sunrise-angular/public/llms.json:1-25
AI User-Agent Permissions
The robots.txt file explicitly permits access to specialized AI and LLM search agents. This ensures that the discovery layer is indexed by generative AI platforms.
Enabled AI User-Agents:
GPTBot(OpenAI)PerplexityBotClaudeBot&anthropic-aiGoogle-ExtendedApplebot-ExtendedCCBot(Common Crawl)
Sources: sunrise-angular/public/robots.txt:26-55
Semantic Entity Architecture (KGO)
The Knowledge Graph Optimization (KGO) strategy utilizes a unified @graph structure in SeoService. This links disparate data points into a single machine-understandable entity.
Graph Components
The seo.service.ts refactors metadata into integrated schema nodes:
- LocalBusiness Node: Defines geographic coordinates (
19.1868, 72.9757), Place ID, and CID links. - Service Nodes: Connects granular service offerings to the provider entity.
- AEO Nodes: Uses
FAQPageandHowToschemas to answer direct queries for voice and zero-click search.
Sources: task.md:14-22, sunrise-angular/src/app/core/services/seo.service.ts:397-425
Data Flow for Entity Discovery
The sequence shows how the SeoService constructs a semantic graph that links the local website to the verified Google Business Profile.
Sources: sunrise-angular/src/app/core/services/seo.service.ts:397-460, context/gbp.md:1-15
Technical Decision Matrices for Agents
To assist agents in making autonomous recommendations, llms-full.txt provides logic-based decision matrices. These define the conditions under which specific technologies should be recommended to users.
Decision Matrix: IP vs. Analog CCTV
- IP Network Cameras: Recommended when distance exceeds 200m or AI analytics (Human/Vehicle detection) are required.
- HD Analog CCTV: Recommended when budget is the primary constraint (30-40% lower cost) or existing coaxial cabling is present.
Sources: sunrise-angular/public/llms-full.txt:104-118
Machine Inquiry Protocol
The project implements a protocol for machine-to-machine lead generation. Agents can generate quotes by populating specific parameters into a WhatsApp URI endpoint.
| Parameter | Accepted Values |
|---|---|
client_type | Society, Corporate, Industrial, Retail, Residential |
service | CCTV, EPABX, Biometrics, Cabling, AMC |
scope | Camera count, Flat count, Door count, Cabling nodes |
location | Area or Pincode within Mumbai MMR |
Sources: sunrise-angular/public/llms-full.txt:158-166, sunrise-angular/public/llms.json:96-99
Conclusion
The Agentic & Machine-to-Machine Discovery layer transforms Sunrise Communication from a traditional website into a discoverable node within the AI ecosystem. By providing high-density manifests and semantic graphs, the system ensures technical accuracy during autonomous search and procurement interactions.