Borrowing it
Nothing to install: this file belongs to mckinsey/agents-at-scale-ark. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/mckinsey/agents-at-scale-ark/main/.claude/skills/ark-controller-development/SKILL.mdgit clone --depth 1 https://github.com/mckinsey/agents-at-scale-arkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mckinsey/agents-at-scale-ark/ark-controller-development)<a href="https://agentmods.dev/skills/mckinsey/agents-at-scale-ark/ark-controller-development"><img src="https://agentmods.dev/badge/skills/mckinsey/agents-at-scale-ark/ark-controller-development.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00039 | $0.01699 |
| Opus 5 | $0.00019 | $0.00849 |
| Sonnet 5 | $0.00008 | $0.00340 |
| Haiku 4.5 | $0.00004 | $0.00170 |
Grade A, and why
ark-controller-development scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
func fetch(ctx, url) (*Metadata, error) { /* 30 lines */ } How it starts
The opening of the file, as written. The whole thing — 194 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ark Controller Development
Guidance for developing the Ark Kubernetes operator in ark/.
When to use this skill
- Modifying Go type definitions (
api/v1alpha1/*_types.go) - Fixing CRD/Helm chart sync errors
- Adding new CRD fields or resources
CRD Generation Flow
api/v1alpha1/*_types.go # Go types with markers
↓
make manifests # Generates CRDs and syncs to Helm chart
↓
config/crd/bases/*.yaml # Source CRDs (auto-generated)
dist/chart/templates/crd/ # Helm chart CRDs (auto-synced)
make manifests automatically syncs source CRDs to the Helm chart while preserving templated headers.
Fixing "CRDs out of sync" Errors
When make build fails with CRD validation errors:
cd ark
make manifests
make build
Key Directories
| Directory | Purpose |
|---|---|
api/v1alpha1/ |
Go type definitions |
config/crd/bases/ |
Auto-generated source CRDs |
dist/chart/templates/crd/ |
Helm chart CRDs (auto-synced) |
internal/controller/ |
Reconciliation logic |
internal/webhook/ |
Admission webhooks |
internal/genai/ |
AI/ML execution logic |
Common Tasks
After Modifying Types or Comments
Go type comments become CRD field descriptions:
cd ark
make manifests
make build
After Any Go Code Change
make lint-fix # Format and auto-fix what the linter can fix
make lint # Fail fast on issues CI will reject — MANDATORY
make build # Build and validate
make test # Run unit + envtest suites
Before Opening a PR
Run the chainsaw e2e suite against a live cluster — unit + envtest don't exercise the full apply → reconcile → status loop. Prefer the deterministic mock-llm suite; only run the LLM-backed tests when the change specifically touches that path (saves cost + avoids flakes from real LLM providers).
# Deterministic suite (default)
(cd tests && chainsaw test --selector '!llm')
# LLM suite — only when your change needs it
(cd tests && chainsaw test --selector 'llm')
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 194 lines · 39 tokens per session scan A 3e522835728e
ark-controller-development is a skill published in the GitHub repository mckinsey/agents-at-scale-ark (422 stars, last pushed today), licensed Apache-2.0. It adds 39 tokens to every session and 1,699 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
new-terraform-provider
Use this when scaffolding a new Terraform provider with the Plugin Framework: workspace layout, go module setup, provider server main.go, and a provider.go with schema and Configure. Also use when a user wants to start building a provider for a new API or asks how to begin a terraform-provider- project.
migration
A step-by-step guide for upgrading programming languages, frameworks, and libraries while checking for breaking changes. A breaking change is an update that requires existing code to be modified.
go
Use when writing, reviewing, testing, or shipping Go code and HTTP services: idioms, %w error wrapping, goroutine/context/errgroup concurrency, net/http 1.22 routing, log/slog, project layout, table-driven tests, Go hardening. NOT language-agnostic threat modeling (that is secure-coding), NOT Dockerfile/CI shipping…
modal
Use when running Python or GPU workloads serverlessly on Modal — modal.App, inline container Images, gpu= on @app.function, Volumes for weight caching, Cron schedules, ASGI endpoints, modal run vs serve vs deploy. NOT managed prediction APIs with no container of your own (that is replicate); NOT SSH-able GPU boxes…
geoserver-cloud
Use when deploying GeoServer on Kubernetes as cloud-native microservices with auto-scaling, service discovery, and centralized configuration. GeoServer Cloud: break monolithic GeoServer into independently scalable WMS/WFS/WCS services.
performance
A guide to measuring and improving the speed and resource use of Go programs. It explains benchmarks and profiling, which records where a program spends time or memory.