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 skills add summarybotng/summarybot-ng --skill shift-right-testinggit clone --depth 1 https://github.com/summarybotng/summarybot-ngWrote 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/summarybotng/summarybot-ng/shift-right-testing)<a href="https://agentmods.dev/skills/summarybotng/summarybot-ng/shift-right-testing"><img src="https://agentmods.dev/badge/skills/summarybotng/summarybot-ng/shift-right-testing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/summarybotng/summarybot-ng/shift-right-testing"><img src="https://agentmods.dev/badge/skills/summarybotng/summarybot-ng/shift-right-testing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00032 | $0.01581 |
| Opus 5 | $0.00016 | $0.00790 |
| Sonnet 5 | $0.00006 | $0.00316 |
| Haiku 4.5 | $0.00003 | $0.00158 |
Grade A, and why
shift-right-testing 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 5d 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.
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 — 234 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Shift-Right Testing
<default_to_action> When testing in production or implementing progressive delivery:
- IMPLEMENT feature flags for progressive rollout (1% → 10% → 50% → 100%)
- DEPLOY with canary releases (compare metrics before full rollout)
- MONITOR with synthetic tests (proactive) + RUM (reactive)
- INJECT failures with chaos engineering (build resilience)
- ANALYZE production data to improve pre-production testing
Quick Shift-Right Techniques:
- Feature flags → Control who sees what, instant rollback
- Canary deployment → 5% traffic, compare error rates
- Synthetic monitoring → Simulate users 24/7, catch issues before users
- Chaos engineering → Netflix-style failure injection
- RUM (Real User Monitoring) → Actual user experience data
Critical Success Factors:
- Production is the ultimate test environment
- Ship fast with safety nets, not slow with certainty
- Use production data to improve shift-left testing </default_to_action>
Quick Reference Card
When to Use
- Progressive feature rollouts
- Production reliability validation
- Performance monitoring at scale
- Learning from real user behavior
Shift-Right Techniques
| Technique | Purpose | When |
|---|---|---|
| Feature Flags | Controlled rollout | Every feature |
| Canary | Compare new vs old | Every deployment |
| Synthetic Monitoring | Proactive detection | 24/7 |
| RUM | Real user metrics | Always on |
| Chaos Engineering | Resilience validation | Regularly |
| A/B Testing | User behavior validation | Feature decisions |
Progressive Rollout Pattern
1% → 10% → 25% → 50% → 100%
↓ ↓ ↓ ↓
Check Check Check Monitor
Key Metrics to Monitor
| Metric | SLO Target | Alert Threshold |
|---|---|---|
| Error rate | < 0.1% | > 1% |
| p95 latency | < 200ms | > 500ms |
| Availability | 99.9% | < 99.5% |
| Apdex | > 0.95 | < 0.8 |
Feature Flags
// Progressive rollout with LaunchDarkly/Unleash pattern
const newCheckout = featureFlags.isEnabled('new-checkout', {
userId: user.id,
percentage: 10, // 10% of users
allowlist: ['beta-testers']
});
if (newCheckout) {
return <NewCheckoutFlow />;
} else {
return <LegacyCheckoutFlow />;
}
// Instant rollback on issues
await featureFlags.disable('new-checkout');
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 5d ago First seen · 234 lines · 32 tokens per session scan A 899fbfd62eba
shift-right-testing is a skill published in the GitHub repository summarybotng/summarybot-ng (2 stars, last pushed 3mo ago), licensed MIT. It adds 32 tokens to every session and 1,581 once invoked, about $0.0002 per session on Opus 5. 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 2026-09-03.
Other skills, from other repositories
ruview-verify
Verify a RuView build — full Rust workspace tests, the deterministic Python pipeline proof (SHA-256 Trust Kill Switch), firmware hash manifest, and the ADR-028 witness bundle with one-command self-verification. Use after any significant change, before merging a PR, or to produce an attestation bundle for a recipient.
evolve
Evolve this harness with Darwin Mode — frozen model, evolving harness (real, sandboxed, safety-gated).
langsmith-observability
LLM observability platform for tracing, evaluation, and monitoring. Use when debugging LLM applications, evaluating model outputs against datasets, monitoring production systems, or building systematic testing pipelines for AI applications.
train-pose
Train/evaluate WiFi pose models honestly — camera-supervised (MediaPipe + CSI) and camera-free (WiFlow), always checked against the mean-pose baseline before any PCK is quoted.
verify-homecore
Run the smallest relevant core, Wasmtime, HAP, or full Homecore test profile.
lovable
Lovable MCP tool usage — deploy projects, message agent, test with agent-browser. Minimize credits by delegating to local repo.