# Perplexity Integration

Canonical: https://www.aiventic.ai/integrations/perplexity

Your service experience, inside Perplexity

Perplexity is good at finding what’s been published about a piece of equipment. Aiventic adds what your own technicians have learned fixing it: similar repairs, the fixes that didn’t hold, and the parts that solved it.

- Category: Agent Harness
- Status: Available

## What syncs

- find_similar_repairs: past jobs on the same model and symptom, and which fixes held
- get_failure_patterns: what tends to fail on a model, and when
- predict_required_parts: the parts a job is likely to need, before dispatch
- prepare_service_call: likely cause, parts, questions to ask, and the best tech for the job

## What Aiventic adds

### Answers from your own jobs

Perplexity knows equipment in general. Through Aiventic it answers from the repairs your company has closed, with the evidence behind each answer.

### Scoped and logged

OAuth for MCP clients and scoped API keys for server-to-server calls. Each agent gets only the tools you allow, and every call is logged.

### Your data stays yours

Your service history isn’t used to train models for anyone else.

### Pay for what agents use

Usage-based: each tool call is metered, and no seat plan is required.

## FAQ

### How do we connect Perplexity?

Connect Perplexity to the Aiventic MCP server and sign in, or call the REST API from your own Perplexity-based tooling.

### What can Perplexity do with Aiventic?

Look up similar past repairs, failure patterns by model, and the parts past jobs used, and prepare a service call with the likely cause, parts, and questions to ask. Everything comes from your own service history and documentation.

### How is it priced?

By usage. Each MCP tool call or API request is metered, and you don’t need a seat plan. Contact us for current rates.

### Is our data used to train models?

No. Your service history isn’t used to train models for anyone else, and each agent only reaches the tools you allow.

---

*Markdown representation of [https://www.aiventic.ai](https://www.aiventic.ai). Append `.md` to any page URL, or send `Accept: text/markdown`, for its markdown. Full site index: [/llms.txt](https://www.aiventic.ai/llms.txt) · [/sitemap.xml](https://www.aiventic.ai/sitemap.xml)*