# Argocd-AI-Assistant

> Source: <https://github.com/saidsef/argocd-ai-assistant>
> Published: 2026-07-12 23:00:39+00:00

An Argo CD UI extension that adds an **Assistant** tab to the resource view. Open any Kubernetes resource, ask questions about it in natural language, and the extension enriches each query with context:

**Resource manifest**- the live object, provided by Argo CD.** Events**- fetched from the Argo CD API and attached automatically.** Logs**(optional) - from a single container, via a guided flow (Pod, Deployment, StatefulSet, Job, Rollout).

A single generic provider speaks the **OpenAI-compatible chat completions API**, so it works with any backend that does: a local inference server (e.g. Ollama), [vLLM](https://github.com/vllm-project/vllm), [OpenAI](https://openai.com/), [Azure OpenAI](https://azure.microsoft.com/en-us/products/ai-services/openai-service/), or any other OpenAI-compatible endpoint. Traffic is routed through the Argo CD [Proxy Extension](https://argo-cd.readthedocs.io/en/stable/developer-guide/extensions/proxy-extensions/) to avoid CORS.

[Argo CD](https://argo-cd.readthedocs.io/en/stable/)>= v2.13- A running LLM backend with an OpenAI-compatible API accessible from the Argo CD server

```
# Install dependencies (requires --force due to React peer dependency quirks)
yarn install --force

# Production build
yarn run build

# Development build
yarn run build-dev
```

Releases are automated: on merge to `main`

(once CI passes), a workflow classifies the change, bumps the version, tags it, and publishes a GitHub Release with the extension tar. No manual version bumping. See [Build and Package](/saidsef/argocd-ai-assistant/blob/main/docs/deployment.md#build-and-package) for details.

See the [Deployment Guide](/saidsef/argocd-ai-assistant/blob/main/docs/deployment.md) for step-by-step instructions on building, packaging, hosting, and installing the extension into Argo CD via the Extension Installer. It covers the Argo CD Operator, the community Helm chart, and raw manifests.

To try the extension end to end on a throwaway [kind](https://kind.sigs.k8s.io/) cluster - with a built-in mock LLM, no GPU or API key required - use the harness in [ examples/kind/](/saidsef/argocd-ai-assistant/blob/main/examples/kind):

```
./examples/kind/setup.sh raw    # or: helm | operator
```

A successful run installs Argo CD, installs this extension built from source, and verifies the full path (including a streamed proxy request) with `== 8 passed, 0 failed ==`

. See [ examples/kind/README.md](/saidsef/argocd-ai-assistant/blob/main/examples/kind/README.md).

Full documentation is in the [ docs/](/saidsef/argocd-ai-assistant/blob/main/docs) directory and built with

[MkDocs Material](https://squidfunk.github.io/mkdocs-material/).

Live docs: [argocd-ai-assistant.readthedocs.io](https://argocd-ai-assistant.readthedocs.io/)

Source and releases: [github.com/saidsef/argocd-ai-assistant](https://github.com/saidsef/argocd-ai-assistant).

Pull requests welcome - see the [Contribution Guide](/saidsef/argocd-ai-assistant/blob/main/CONTRIBUTING.md).