> ## Documentation Index
> Fetch the complete documentation index at: https://docs.neuralfactory.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Chat Skill

> Stream multi-turn AI-powered skill creation using a ReAct agent with tools.

Accepts either a JSON body ({"message", "thread_id"}) or multipart/form-data
with the same fields plus optional ``files`` attachments. Attachments are
validated up front (type/size/count) and inlined into the agent message as
text blocks — text files verbatim, document formats via text extraction.

When ``skill_id`` is supplied (and the caller is a user), the turn is
persisted to the caller's per-skill chat thread: the user message (or the
compact ``display_message``) up front, the assistant result in the stream's
cleanup once it produced output.

API key permission: Skills (Admin also opens it).



## OpenAPI

````yaml /api-reference/openapi.json post /skills/chat
openapi: 3.1.0
info:
  title: Neural Factory API
  version: 0.1.0
  description: >-
    The endpoints a program can call with an API key. Exchange the key's secret
    for a token at POST /apikeys/token and send it as a Bearer token.
servers:
  - url: https://api.neuralfactory.ai
    description: SaaS
  - url: https://{apiHost}
    description: 'Enterprise: the API address of your own deployment'
    variables:
      apiHost:
        default: api.example.com
        description: Your deployment's API host
security:
  - bearerAuth: []
tags:
  - name: API keys
  - name: Chat
  - name: Runs
  - name: Agents
  - name: Workflows
  - name: Schedules
  - name: Knowledge bases
  - name: Documents
  - name: Search
  - name: Skills
  - name: Toolkits
  - name: Files
  - name: Interactions
  - name: Usage and billing
  - name: Catalogue
  - name: Browser
  - name: Users
  - name: Groups
  - name: Builds
paths:
  /skills/chat:
    post:
      tags:
        - Skills
      summary: Chat Skill
      description: >-
        Stream multi-turn AI-powered skill creation using a ReAct agent with
        tools.


        Accepts either a JSON body ({"message", "thread_id"}) or
        multipart/form-data

        with the same fields plus optional ``files`` attachments. Attachments
        are

        validated up front (type/size/count) and inlined into the agent message
        as

        text blocks — text files verbatim, document formats via text extraction.


        When ``skill_id`` is supplied (and the caller is a user), the turn is

        persisted to the caller's per-skill chat thread: the user message (or
        the

        compact ``display_message``) up front, the assistant result in the
        stream's

        cleanup once it produced output.


        API key permission: Skills (Admin also opens it).
      operationId: chat_skill_skills_chat_post
      parameters:
        - name: env
          in: query
          required: false
          schema:
            title: Env
      responses:
        '200':
          description: Successful Response
          content:
            application/json:
              schema: {}
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
components:
  schemas:
    HTTPValidationError:
      properties:
        detail:
          items:
            $ref: '#/components/schemas/ValidationError'
          type: array
          title: Detail
      type: object
      title: HTTPValidationError
    ValidationError:
      properties:
        loc:
          items:
            anyOf:
              - type: string
              - type: integer
          type: array
          title: Location
        msg:
          type: string
          title: Message
        type:
          type: string
          title: Error Type
        input:
          title: Input
        ctx:
          type: object
          title: Context
      type: object
      required:
        - loc
        - msg
        - type
      title: ValidationError
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      bearerFormat: JWT
      description: >-
        The access_token returned by POST /apikeys/token for an API key's
        secret.

````

This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.