lenaxia

lenaxia / k8s-mendabot

Public

A K8s controller that watches your cluster for failures and opens pull requests on your GitOps repository with fixes. Security is a first class citizen, and it runs in-cluster with read-only RBAC, redacts secrets before they reach LLM, and requires human-in-the-loop. Formerly known as k8s-mendabot

agent argocd automation cloud-native devops
13
0
100% credibility
Found Feb 28, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

A Kubernetes operator that monitors for pod crashes, deployment issues, and node failures, then uses an in-cluster AI agent to diagnose problems and create pull requests with fixes in your GitOps repository.

How It Works

1
🔍 Discover Mendabot

You hear about a smart helper that watches your cluster for problems and suggests fixes.

2
🔑 Prepare connections

You link your code storage account and an AI thinking service so it can read your setups and get smart help.

3
🚀 Install with one click

You add it to your cluster easily, and it starts running quietly in the background.

4
📁 Tell it your config spot

You show it where your cluster setup files live so it knows what to check and change.

5
🛡️ It guards your cluster

Now it's watching everything, ready to jump in when trouble starts.

6
🚨 Spot a problem

A pod crashes or something goes wrong, and it notices right away.

7
🤖 AI investigates automatically

It sends a smart agent inside the cluster to figure out what's wrong and how to fix it.

✅ Get a ready-to-merge fix

You receive a pull request with the exact change needed—review, approve, and your cluster heals itself.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 13 to 13 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is k8s-mendabot?

k8s-mendabot is a Go-based Kubernetes controller that monitors your cluster for failures like crash-looping pods, degraded deployments, or unhealthy nodes, then spawns an in-cluster LLM agent to diagnose issues and open pull requests with fixes directly in your GitOps repo. Using the k8s controller pattern and controller-runtime, it detects problems natively via the API, deduplicates alerts by parent resource, and enforces a stabilization window before acting—all with read-only RBAC and secret redaction before hitting any OpenAI-compatible endpoint, including self-hosted models. Install via Helm, configure your GitHub App and LLM key, and it handles the rest with human review required.

Why is it gaining traction?

It stands out in the k8s controller vs operator debate by staying lightweight and fully in-cluster—no external databases or services—while prioritizing security as a first-class citizen with prompt injection defenses, network policies, and short-lived GitHub tokens. Developers love the structured PRs (summary, evidence, fix, confidence score) that integrate seamlessly with GitOps tools like Flux or ArgoCD, plus Prometheus metrics and per-resource annotations for fine control. For github k8s at home or monitoring setups, it's a practical k8s controller example that automates tedious debugging without risking your cluster.

Who should use this?

SREs running production GitOps clusters who want automated incident triage before pages fire. K8s github actions runner maintainers tired of manual pod fixes. Hobbyists with github k8s at home stacks seeking self-healing without external SaaS.

Verdict

Solid early prototype (12 stars, 1.0% credibility) with exceptional docs and security hygiene—try it in dev clusters via Helm for k8s controller reconcile workflows, but wait for more battle-testing before prod. Pairs well with k8s github monitoring dashboards.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.