Build AI workflows that actually work in production.

Awee Skills is a workflow engine that makes AI automation composable, observable, and cost-accountable - from a single YAML file to a production-grade pipeline.

Awee Skills Engine

The core workflow engine. Define steps in YAML, chain AI calls, manage cost, and run anything - locally or in production.

Awee Platform

The hosted layer - managed runs, dashboards, billing, and team access. Coming soon.

Most workflow tools make you choose: low-code simplicity or engineering flexibility. Awee Skills gives you both. Workflows are plain YAML - readable, diffable, reviewable. The engine is production infrastructure - typed, traced, retried, and cost-metered on every run.

AI inference, first-class

Call OpenAI, Anthropic, Ollama, or OpenRouter from any step. Extended thinking, tool use, structured output. Switch providers by changing one field.

Template everything

Every field in every step is a live expression. Chain step outputs, transform values, apply conditionals - without writing any glue code.

Know what it costs

Per-token pricing in your currency, updated daily. Input, output, and thinking tokens costed separately. Full I/O and timing metrics on every step.

50+ built-in components

Web search, crawling, HTTP, file I/O, CSV, JSON, XML, databases, shell, browser scraping, events, and more - all wired and ready.

A workflow in 30 lines

id: research-and-summarise
name: Research and Summarise

actions:
  - name: search
    component: search:web
    vars:
      query: "{{ topic }} latest developments 2025"

  - name: fetch
    component: crawler
    vars:
      url: "{{ search.results[0].url }}"

  - name: summarise
    component: inference
    vars:
      provider: anthropic
      model: claude-opus-4-20250514
      prompt: |
        Summarise the following article in three bullet points.
        Article: {{ fetch.output }}
    cache:
      for: 1h

  - name: save
    component: file:write
    vars:
      path: "./output/{{ topic | slug }}.md"
      content: "{{ summarise.content }}"

Search the web, crawl the top result, ask an AI to summarise it, write to a file. The summary is cached for one hour - re-running with the same inputs skips the model call entirely. That's the whole workflow.