Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/skyhook-io/radar/visual-testgit clone --depth 1 https://github.com/skyhook-io/radarWhat 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 | $0.00000 | $0.02332 |
| Opus 5 | $0.00000 | $0.01166 |
| Sonnet 5 | $0.00000 | $0.00466 |
| Haiku 4.5 | $0.00000 | $0.00233 |
Grade A, and why
visual-test scanned grade A with 0 findings 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 yesterday.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Visual Test
Visually test Radar changes against a real cluster using Playwright MCP. Takes screenshots, organizes them, and reports what looks right or wrong.
Step 1: Verify Playwright MCP
Check if Playwright MCP tools are available (e.g., mcp__playwright__browser_navigate). If NOT available, tell the user "Playwright MCP is not available -- please add it to your MCP config and restart" and stop.
Step 2: Understand What to Test
Determine what changed and what needs visual verification:
git status
git branch --show-current
git log main..HEAD --oneline
git diff main..HEAD --name-only
If on main with no branch commits, check unstaged changes instead. Summarize what changed and what UI areas are affected (which renderers, which views, which components).
Step 3: Verify Cluster Access
This is critical -- Radar is useless without a live cluster. Check in order:
# 1. Check current kubeconfig context
kubectl config current-context
# 2. Verify actual connectivity (not just config)
kubectl cluster-info --request-timeout=5s
If cluster is not reachable, try to diagnose:
# GKE:
gcloud container clusters list 2>/dev/null | head -5
# EKS:
aws eks list-clusters 2>/dev/null | head -5
If auth is expired or cluster is unreachable:
- Tell the user which context is configured and that it's not reachable
- Suggest the auth command (e.g.,
gcloud container clusters get-credentials ...oraws eks update-kubeconfig ...) - Ask them to run
! <auth-command>in the prompt to authenticate - Stop and wait -- don't proceed without a working cluster
Step 4: Check Required Resources Exist
Based on what changed, verify the cluster has the CRD resources needed to test:
# Check if the CRD exists
kubectl api-resources --verbs=list 2>/dev/null | grep -i <kind>
# Check if instances exist
kubectl get <kind> -A --no-headers 2>/dev/null | head -5
If CRDs exist but no instances:
- Tell the user: "The cluster has the CRD but no instances. Want me to create a test resource?"
- Wait for confirmation before creating anything
- If creating: use a clearly labeled test namespace/name (e.g.,
radar-test/test-<kind>)
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.
- yesterday First seen · 195 lines · 0 tokens per session scan A cdd534660b52
visual-test is a command published in the GitHub repository skyhook-io/radar (3,168 stars, last pushed yesterday), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 2,332 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
OPSX: Fast Forward
Create a change and generate all artifacts needed for implementation in one go.
OPSX: Apply
Implement tasks from an OpenSpec change (Experimental).
OPSX: Propose
Propose a new change - create it and generate all artifacts in one step.
OPSX: Sync
Sync delta specs from a change to main specs.
OPSX: Onboard
Guided onboarding - walk through a complete OpenSpec workflow cycle with narration.
OPSX: Archive
Archive a completed change in the experimental workflow.