The Flux CLI now supports plugins and we have two official ones you can check out:
- Flux Mirror for syncing Helm charts, OCI artifacts and container images between registries.
- Flux Schema for validating Kubernetes manifests against JSON schemas and CEL rules, offline from clusters.
These plugins should be useful to anyone working with OCI and Kubernetes.
Notably flux-mirror uses a CNCF license which makes it a great replacement for bitnami/charts-syncer.
This release also adds:
- Field ignore rules that let server-side apply share ownership with other controllers.
- Git commit signing and verification using SSH keys.
- Kubernetes Workload Identity authentication for OpenBao/Vault and AWS CodeCommit.
- Post-Quantum SOPS decryption for secrets using compatible age ciphers.
In the ecosystem, the Flux Operator v0.53.0 web UI has a new Workloads Dashboard and can view logs for Pods.
hi friends, Flux maintainer here. Stefan gave this talk at FluxCon EU
He maps out some "levels of detachment" for software & platform engineering that we've all been watching happen. Copilot autocomplete -> chat interfaces -> spec-driven development -≥ agent teams -≥ fully autonomous dark factories. Same progression applies to ops: agents that go into your clusters, diagnose issues, and present resolution plans without human intervention.
GitOps still matters in this world of agents:
1. Git as the interface to production (agents open PRs instead of running kubectl directly)
2. Git as persistent memory (incident RCA gets committed so the next similar event resolves faster)
3. Desired state that includes agent definitions (recover your AI orchestration the same way you recover a cluster: point Flux at a repo)
He also announced fluxcd/agent-skills, which are OCI packaged, verifiable skills for Flux that should progressively discovery instead of loading everything into context. We've seen a 32% improvement on knowledge tasks and 29% on repo audits vs. baseline models running against real public repos.
Stefan's anecdotes on developing agent tooling are relevant beyond Flux.
Every harness defines agents differently (markdown, TOML, JSON), there's no portability, and reproducibility and measurement is expensive and time-consuming.