Kdeps lets you build AI agent workflows in YAML and run them anywhere — locally with Ollama or llamafile (no API key needed), or in the cloud with OpenAI, Anthropic, or Groq.
Kdeps is a strong pick if your team builds AI-powered APIs or bots and wants full control over where the code runs. The YAML-first approach removes most boilerplate, and the offline-first design is rare in this space. The tradeoff: you need engineers comfortable with CLIs and container deployments to get value out of it.
Run a local AI coding agent instantly with `kdeps` — uses Ollama for fully offline, private inference
Write workflows in YAML with `requires:` to declare dependencies between steps; no Python glue code needed
Orchestrate multiple agents ("agencies") declaratively — one agent calls another via the `agent:` resource type
Install pre-built components from the registry: scraper, search, browser, embedding, TTS, email
Deploy the same workflow.yaml as Docker, Kubernetes, single binary, or bootable ISO — no format changes
Source: Kdeps·Verified July 2026
No integrations listed yet for Kdeps.
Local AI REPL powered by Ollama or llamafile (no API key required); supports OpenAI, Anthropic, Groq, and any OpenAI-compatible endpoint; LLM decides which tools to call in agent mode; AI skills for Claude Code and Cursor scaffold workflows automatically.
AI-generated training guides tailored to your team's size, skill level, and focus areas for Kdeps — coming in v0.3.2.
View our roadmap →We're building a review system so business owners like you can share real experiences with Kdeps.
Last researched: July 2026
