> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://dev.neurologyca.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://dev.neurologyca.com/_mcp/server.

# Overview

> Compare Neurologyca's SoftSkills, Agents, and Engine APIs.

Neurologyca currently exposes three different API patterns, each built for a different product experience. The right integration depends on whether you want to analyze a recorded video, embed a conversational coaching experience, or process live session data through the Engine pipeline.

#### [SoftSkills API](/softskills-overview)

Upload a recorded video and receive structured communication feedback as JSON.

#### [Agents API](/agents-overview)

Build session-based chat and coaching experiences with traced multi-agent responses.

#### [Engine API](/engine-overview)

Process live-session packages and retrieve Constellation insights after the session completes.

## API options at a glance

| API                | Best for                                              | Integration model                                      | Authentication                                          | Start here                                                 |
| ------------------ | ----------------------------------------------------- | ------------------------------------------------------ | ------------------------------------------------------- | ---------------------------------------------------------- |
| **SoftSkills API** | Recorded presentation or communication analysis       | Upload then immediate result or async polling workflow | API key (`Authorization: Bearer ng_...` or `X-API-Key`) | `POST /api/v1/full-analysis`                               |
| **Agents API**     | Chat, coaching, and guided conversational experiences | Session-based workflow                                 | Bearer token                                            | `POST /api/v1/auth/firebase`, then `POST /api/v1/sessions` |
| **Engine API**     | Live or step-based video session processing           | Multi-step pipeline                                    | Engine session token from `POST /session`               | `POST /session`                                            |

## Which API should you use?

* Choose **SoftSkills API** if your application already has a finished video file and you want Neurologyca to return a full analysis result immediately or through an async job.
* Choose **Agents API** if you want to embed a Neurologyca-powered assistant, preserve chat history, and display multi-agent traces or orchestration details in your UI.
* Choose **Engine API** if your product captures a live assessment session, uploads `.kpn` packages over time, and needs a final Constellation analysis after processing completes.

## Important integration differences

* **Auth is different for each API.** SoftSkills uses API keys, Agents uses bearer auth from your application identity flow, and Engine uses the token returned when you create a session.
* **The request shapes are different.** SoftSkills accepts multipart video uploads, Agents accepts JSON session and message payloads, and Engine accepts both JSON and multipart uploads with `.kpn` packages.
* **The workflow depth is different.** SoftSkills is still the simplest public integration, but it can return either a synchronous result or a saved async session state. Agents is conversational and stateful, and Engine is designed for a staged processing pipeline.

## Next step

If you already know your use case, go directly to the product section you need:

* [SoftSkills Overview](/softskills-overview)
* [Agents Overview](/agents-overview)
* [Engine Overview](/engine-overview)

If you are still deciding, continue to [Getting Started](/getting-started) for a use-case-based guide.