YAMLResume v0.14: AI-Powered Resume Generation

We're excited to announce YAMLResume v0.14. This release brings AI-powered resume generation to YAMLResume with the new yamlresume ai generate command, a brand new @yamlresume/ai package, support for multiple LLM providers, and the usual bug fixes and improvements.

New Package: @yamlresume/ai

At the heart of this release is the new @yamlresume/ai package. It provides AI-powered generation capabilities for YAMLResume and is built on top of the Vercel AI SDK.

The package exposes a clean programmatic API for generating resumes:

import { generateResume } from "@yamlresume/ai";
import { openai } from "@ai-sdk/openai";

const model = openai("gpt-5");

const yaml = await generateResume({
  position: "Software Engineer",
  language: "en",
  model,
});

The generated output is parsed and validated against the YAMLResume schema from @yamlresume/core. If validation fails, the request is retried automatically (default: up to 2 retries) with the previous error fed back to the model, so you get a valid resume YAML instead of a broken draft.

New Command: yamlresume ai generate

The headline feature is the new yamlresume ai generate command, which turns a position and language into a complete, realistic resume file.

Supported providers

YAMLResume supports multiple LLM providers through OpenAI-compatible endpoints. The provider is inferred from environment variables:

The provider is detected in the order shown above. When no provider is detected, Kimi is used as the default so that a missing API key produces a clear configuration error.

Generate your first resume

Set a provider API key and run:

$ export OPENAI_API_KEY=sk-...
$ yamlresume ai generate --position "Software Engineer" --language en resume.yml

The command will generate a complete resume file at resume.yml with sections such as basics, location, profiles, education, work, skills, languages, and relevant optional sections tailored to the position.

Override model and base URL

You can override the model and base URL either via environment variables or CLI flags. CLI flags take precedence over environment variables.

$ yamlresume ai generate \
    --position "Registered Nurse" \
    --language en \
    --model gpt-5 \
    --base-url https://custom.openai.endpoint/v1 \
    resume.yml

Alternatively:

$ export YAMLRESUME_AI_MODEL=gpt-5
$ export YAMLRESUME_AI_BASE_URL=https://custom.openai.endpoint/v1
$ yamlresume ai generate --position "Registered Nurse" --language en resume.yml

Run yamlresume ai generate --help to see all available options and environment variables.

How it works

yamlresume ai generate builds a detailed prompt from the YAMLResume schema rules and a sample resume, then asks the LLM to produce a complete resume. The prompt instructs the model to:

• Output only valid YAML, without markdown fences.
• Keep YAML comments so the generated file remains friendly for editor/IDE validation.
• Follow all schema constraints such as field lengths, date formats, enum values, and valid URLs.
• Invent original, realistic content tailored to the target position and locale, without copying the sample resume's personal details.

The generated text is then parsed and validated. If validation fails, the model is asked to fix the errors and try again, up to the configured retry limit.

Bug Fixes and Improvements

Patch releases since v0.13.0

Between v0.13.0 and v0.14.0 we shipped two patch releases with several improvements:

v0.13.2

• Added the DOCX engine to the default layouts generated by yamlresume new
• Added Discord to the supported social network options
• Extended the maximum length of project descriptions from 128 to 1024 characters
• Added id attributes to HTML section elements for anchor navigation
• Made education.startDate optional

v0.13.1

• Fixed the playground toolbar height and DOCX content scrolling behavior
• Silenced noisy localStorage warnings in the browser console

v0.14.0 fixes

HTML Calm template skill layout

• Skills in the HTML Calm template now stack vertically when using keywords, improving readability on narrow viewports. Credit to couimet for his PR.

Upgrade Today

Get the latest version:

$ npm install -g yamlresume@latest
# or
$ brew upgrade yamlresume

Then generate your first AI resume:

$ yamlresume ai generate --position "Software Engineer" --language en resume.yml

Credit

As always, we welcome your feedback and contributions on GitHub. Happy building!