> ## Documentation Index > Fetch the complete documentation index at: https://docs.vlm.run/llms.txt > Use this file to discover all available pages before exploring further. # Create Skill > Create a skill from a file, prompt, or chat session. Three creation modes (provide exactly one source): A. file_id: Use an already-uploaded skill zip (must contain SKILL.md). B. prompt (+ optional json_schema): Auto-generate SKILL.md and schema.json, package and upload. C. session_id: Retrieve chat history as the prompt, auto-generate SKILL.md. All modes produce a skill zip stored in GCS and deduplicate by content hash. ```python Python theme={"theme":{"light":"github-light","dark":"dark-plus"}} !pip install vlmrun from vlmrun.client import VLMRun client = VLMRun(api_key="") # Create a skill from a prompt skill = client.agent.skills.create( prompt="Extract invoice_id, date, and total_amount from invoices.", json_schema={ "type": "object", "properties": { "invoice_id": {"type": "string"}, "invoice_date": {"type": "string", "format": "date"}, "total_amount": {"type": "number"} }, "required": ["invoice_id", "invoice_date", "total_amount"] } ) print(f"Created skill: {skill.id} ({skill.name})") # Create a skill from a pre-uploaded file skill = client.agent.skills.create( name="invoice-extraction", description="Extract structured data from invoices", file_id="" # Uploaded skill zip containing SKILL.md ) # Create a skill from a chat session skill = client.agent.skills.create( session_id="" ) ``` ```typescript Node.js SDK theme={"theme":{"light":"github-light","dark":"dark-plus"}} import { VlmRun } from "vlmrun"; const client = new VlmRun({ baseURL: "https://api.vlm.run/v1", apiKey: "" }); // Create a skill from a prompt const skill = await client.agent.skills.create({ prompt: "Extract invoice_id, date, and total_amount from invoices.", jsonSchema: { type: "object", properties: { invoice_id: { type: "string" }, invoice_date: { type: "string", format: "date" }, total_amount: { type: "number" } }, required: ["invoice_id", "invoice_date", "total_amount"] } }); console.log(skill); ``` ```bash cURL theme={"theme":{"light":"github-light","dark":"dark-plus"}} # Create from prompt curl -X POST "https://api.vlm.run/v1/skills/create" \ -H "Authorization: Bearer " \ -H "Content-Type: application/json" \ -d '{ "prompt": "Extract invoice_id, date, and total_amount from invoices.", "json_schema": { "type": "object", "properties": { "invoice_id": {"type": "string"}, "invoice_date": {"type": "string", "format": "date"}, "total_amount": {"type": "number"} }, "required": ["invoice_id", "invoice_date", "total_amount"] } }' # Create from pre-uploaded file curl -X POST "https://api.vlm.run/v1/skills/create" \ -H "Authorization: Bearer " \ -H "Content-Type: application/json" \ -d '{ "name": "invoice-extraction", "description": "Extract structured data from invoices", "file_id": "" }' ``` ## OpenAPI ````yaml POST /v1/skills/create openapi: 3.1.0 info: title: VLM Run Unified Server description: Unified server for VLM Run Agent and API termsOfService: https://vlm.run/terms-of-service contact: name: VLM Run Support Team url: https://vlm.run/ email: support@vlm.run version: 2026-05-19.0 servers: [] security: [] paths: /v1/skills/create: post: tags: - skills summary: Create Skill description: >- Create a skill from a file, prompt, or chat session. Three creation modes (provide exactly one source): A. file_id: Use an already-uploaded skill zip (must contain SKILL.md). B. prompt (+ optional json_schema): Auto-generate SKILL.md and schema.json, package and upload. C. session_id: Retrieve chat history as the prompt, auto-generate SKILL.md. All modes produce a skill zip stored in GCS and deduplicate by content hash. operationId: create_skill_v1_skills_create_post requestBody: content: application/json: schema: $ref: '#/components/schemas/SkillCreateRequest' required: true responses: '201': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/SkillCreateResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - HTTPBearer: [] components: schemas: SkillCreateRequest: properties: name: anyOf: - type: string maxLength: 255 - type: 'null' title: Name description: >- Name of the skill (required for file_id, auto-generated for prompt/session_id) description: anyOf: - type: string - type: 'null' title: Description description: >- Description of the skill (required for file_id, auto-generated for prompt/session_id) file_id: anyOf: - type: string - type: 'null' title: File Id description: File ID from file upload API (the uploaded skill zip) prompt: anyOf: - type: string - type: 'null' title: Prompt description: >- Prompt text that becomes the SKILL.md body (auto-generates skill zip) json_schema: anyOf: - additionalProperties: true type: object - type: 'null' title: Json Schema description: JSON schema that becomes schema.json in the skill zip session_id: anyOf: - type: string - type: 'null' title: Session Id description: Chat session ID to retrieve chat history as the prompt is_public: type: boolean title: Is Public description: Whether the skill is publicly accessible default: false type: object title: SkillCreateRequest description: >- Request to create a skill. Three creation modes (provide exactly one): A. file_id: Create from a pre-built skill zip uploaded via POST /v1/files. B. prompt (+ optional json_schema): Auto-generate SKILL.md and schema.json. C. session_id: Retrieve chat history and auto-generate SKILL.md from messages. For mode A (file_id), name and description are required. For modes B (prompt) and C (session_id), name and description are optional — if omitted, they are auto-generated by the LLM from the content. SkillCreateResponse: properties: id: type: string title: Id description: ID of the created skill name: type: string title: Name description: Name of the skill (may differ from request if LLM-generated) description: type: string title: Description description: Description of the skill (may differ from request if LLM-generated) skill_version: type: string title: Skill Version description: Auto-generated version of the skill (e.g. '20260512-abc12345') status: type: string title: Status description: Status of the skill creation default: completed created_at: type: string format: date-time title: Created At description: Date and time when the skill was created updated_at: type: string format: date-time title: Updated At description: Date and time when the skill was updated type: object required: - id - name - description - skill_version - created_at - updated_at title: SkillCreateResponse description: Response for skill creation. 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: HTTPBearer: type: http scheme: bearer ````