GitHub Cost Optimization
OhChimp scans your GitHub repos for oversized Terraform instances, Kubernetes pods missing limits, and heavy Docker images, then writes reviewable fixes.
What OhChimp optimizes for GitHub
Some of the cost levers OhChimp checks for GitHub. Each one becomes a reviewable plan you approve before anything changes.
Oversized instances in Terraform
OhChimp parses your .tf files and pulls out cost-relevant resources like aws_instance, aws_rds, aws_elasticache, google_compute_instance, and azurerm_virtual_machine, along with the instance_type each one declares. That surfaces the sizes pinned in code as candidates to right-size, and OhChimp writes the IaC change so the cheaper size ships and does not drift back.
Kubernetes pods with no resource specs
Manifests that skip CPU and memory requests and limits let pods grab whatever the node has, which inflates the cluster you pay for. OhChimp flags Deployments, StatefulSets, DaemonSets, Jobs, CronJobs, and Pods that are missing requests or limits, and proposes the specs to add so nodes pack tighter.
Heavy Docker base images
A single-stage Dockerfile built on a full base image ships a bigger artifact than it needs, which costs registry storage and slows every build. OhChimp finds single-stage Dockerfiles whose base is not an alpine, slim, or distroless variant, and writes the change.
CI jobs that provision cloud resources
OhChimp scans your .github/workflows files and matches cost patterns in them: cloud provisioning commands like aws ec2 run-instances flag as high severity, and resource sizing and Kubernetes mutations flag as medium. It surfaces the high and medium severity ones so you trim the spend baked into your pipelines.
Provisioning with no teardown
A script that spins up cloud resources but never deletes them leaves them billing long after the job finishes. OhChimp scans your shell scripts, Makefiles, and CI files for create commands that have no matching delete or destroy, and flags the leak so the resources get torn down.
Untracked deploy and registry targets
OhChimp reads your GitHub Actions steps to see where they push and deploy: docker build-push to ghcr, ECR, GCR, or Docker Hub, plus kubectl apply, helm upgrade, and terraform apply. That map ties each repo to the infrastructure it stands up, so cost waste lands on the right team.
GitHub cost optimization FAQ
How does OhChimp connect to GitHub?
Through the OhChimp GitHub App (recommended) or a read-only personal access token. OhChimp reads repository contents and metadata to scan your infrastructure-as-code, Dockerfiles, Kubernetes manifests, and CI workflows. It never stores application data, secrets, or workload contents.
Does OhChimp need write access to my repos?
No. The connection is read-only. OhChimp reads your code to find cost anti-patterns and writes each fix as a reviewable plan. Nothing in your repos or infrastructure changes until you click apply, and you are always the one who clicks.
What costs does scanning my GitHub repos actually reduce?
Oversized instance types hardcoded in Terraform, Kubernetes pods missing CPU and memory requests or limits, heavy Docker base images, CI jobs that provision cloud resources, and provisioning scripts that never tear down what they create. OhChimp also maps where your workflows deploy and push images so waste lands on the right team.
Who applies the changes, and can I roll them back?
You do. Every fix is a plan with a confidence score, a risk level, and rollback steps. You review it, apply it with one click, and can roll it back using the steps in the plan.
How are the savings verified, and what does it cost?
OhChimp checks your actual cloud bill after you apply a plan. A saving is marked VERIFIED only after a drop of at least 10% that holds for 7 or more days across 3 consecutive checks, otherwise it is flagged "not implemented." Pricing is a flat monthly fee with no cut of your savings, plus a year-one ROI guarantee: a full refund of subscription fees if it does not pay for itself in your first 12 months on a paid plan.
Why the oversized-instance lever needs Terraform fixed too: Stop Terraform drift from reverting your right-sizing.
Related integrations
Teams running GitHub usually run these too. OhChimp finds the waste in each and proves it on the bill.
Kubernetes
Pod rightsizing, missing limits, node packing
PostgreSQL
Unused indexes, bloat, and cache efficiency
MySQL
Unused indexes, slow queries, instance sizing