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 e-t-y-b/etyb-skills --skill cloudflaregit clone --depth 1 https://github.com/e-t-y-b/etyb-skillsWrote 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/e-t-y-b/etyb-skills/cloudflare)<a href="https://agentmods.dev/skills/e-t-y-b/etyb-skills/cloudflare"><img src="https://agentmods.dev/badge/skills/e-t-y-b/etyb-skills/cloudflare/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/e-t-y-b/etyb-skills/cloudflare"><img src="https://agentmods.dev/badge/skills/e-t-y-b/etyb-skills/cloudflare.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 | $0.00262 | $0.03965 |
| Opus 5 | $0.00131 | $0.01982 |
| Sonnet 5 | $0.00052 | $0.00793 |
| Haiku 4.5 | $0.00026 | $0.00396 |
Grade A, and why
stack-cloudflare 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 12d 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.
- **Workers RPC is the canonical inter-worker pattern.** Service bindings that did `env.OTHER.fetch(request)` are legacy. Modern bindings use `WorkerEntrypoint` classes and method calls (`await env.OTHER.someMethod(args) How it starts
The opening of the file, as written. The whole thing — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cloudflare Stack — Team Briefing
This is a knowledge overlay, not a new specialist. The existing ETYB team does the work — backend-architect writes the backend code, devops-engineer wires the deploys, security-engineer enforces the boundary. This pack tells each role where the current Cloudflare knowledge lives.
Where the full briefing lives
The full Stack briefing lives in this same folder. Per-product and per-role pages are siblings of this SKILL.md. Every page carries last_verified_on stamps and authoritative-source URLs in its frontmatter; see skills/etyb/core/knowledge-currency.md for the drift-check protocol that uses them.
- Stack briefing:
stacks/cloudflare/index.md - Per-product pages:
stacks/cloudflare/<product>.md— one per entry inproducts_coveredabove - Per-role views:
stacks/cloudflare/<role>.md— one per role inapplies_to_rolesabove
When ETYB is installed locally these are read directly from disk. For third-party agents without the install, the same content is reachable as raw markdown at https://raw.githubusercontent.com/e-t-y-b/etyb-skills/main/stacks/cloudflare/<page>.md.
When delegate_to_skills (frontmatter above) lists a first-party vendor MCP/skill that's installed in the user's environment, ETYB defers to it first. The in-repo Stack content is the curated fallback.
What changed in 2025-2026 that older training data misses
Critical context — an LLM with a 2024 cutoff will get these wrong:
- Workers RPC is the canonical inter-worker pattern. Service bindings that did
env.OTHER.fetch(request)are legacy. Modern bindings useWorkerEntrypointclasses and method calls (await env.OTHER.someMethod(args)). RPC traverses Workers, Durable Objects, and Workers for Platforms — same primitive. - Durable Objects are SQLite-backed by default. New DO classes get a per-DO SQLite database, not the legacy KV-style storage. Transactions, full SQL, alarms, and ~10GB per DO are standard. Old DO storage API (
this.state.storage.put/get) still works but new code should use SQL. - D1 supports global read replication and the Sessions API. Multi-region read replicas as of 2025; the Sessions API ensures read-your-writes consistency across replicas. Don't build D1 apps that assume a single primary location anymore.
- Hyperdrive supports MySQL in addition to Postgres, and can target private databases over Cloudflare Tunnel. It is now the standard way to put a relational DB behind Workers — connection pooling + query caching + private connectivity in one binding.
- Vectorize V2 is the current generation: up to 1536-dim (and larger for some plans), metadata indexes for filtered search, multi-million-vector indexes per account, namespace partitioning. v1 indexes still exist but new builds should target V2.
- Workflows GA'd in 2025. Durable-execution primitive (think Temporal-as-a-binding). Long-running, retry-aware, sleepable. New mental model for orchestrating async work — don't reach for Queues + state-in-DO when Workflows fit.
- Pages is in maintenance mode. New projects should use Workers Static Assets with the
assetsbinding. Pages keeps working; new platform features land in Workers Static Assets first. - AutoRAG was renamed to Cloudflare AI Search in 2025. Same product (managed RAG pipeline on R2 + Vectorize + Workers AI).
- Realtime (TURN + SFU + Realtime API) is GA. Replaces the older "Cloudflare Calls" naming. Real-time audio/video apps on Workers without managing your own SFU.
- Containers on Workers (beta through 2025) lets a Worker spawn a long-running container for heavy or stateful workloads — bridges the gap to Node/Python apps that don't fit the V8-isolate model.
- Wrangler v4 is current.
wrangler deploy(notpublish),wrangler dev(with --remote),wrangler tail,wrangler types,wrangler secret bulk. Oldwrangler publishand Wrangler v1/v2 syntax must be flagged and replaced. compatibility_date+compatibility_flagsare mandatory. Without them the runtime falls back to old behavior. New projects should pincompatibility_datenear current and setnodejs_compatornodejs_compat_v2if needed.
What ships with it
45 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.
- access.md 5.9 KB
- ai-gateway.md 6.1 KB
- ai-ml-engineer.md 15 KB
- ai-search.md 4.1 KB
- analytics-engine.md 5.6 KB
- argo.md 4.1 KB
- backend-architect.md 17 KB
- browser-rendering.md 4.1 KB
- cron-triggers.md 3.8 KB
- d1.md 8.7 KB
- database-architect.md 17 KB
- ddos.md 4.9 KB
- devops-engineer.md 13 KB
- durable-objects.md 7.8 KB
- email-routing.md 3.1 KB
- email-workers.md 6.0 KB
- hyperdrive.md 5.7 KB
- images.md 3.4 KB
- index.md 17 KB
- kv.md 4.6 KB
- logpush.md 5.9 KB
- magic-transit.md 5.0 KB
- pages.md 4.0 KB
- pipelines.md 4.0 KB
- queues.md 5.3 KB
- r2.md 6.2 KB
- rate-limiting.md 5.5 KB
- realtime.md 4.2 KB
- security-engineer.md 19 KB
- smart-placement.md 4.5 KB
- stream.md 3.4 KB
- system-architect.md 16 KB
- tunnel.md 6.5 KB
- turnstile.md 5.6 KB
- vectorize.md 5.8 KB
- waf.md 5.8 KB
- workers-ai.md 7.2 KB
- workers-for-platforms.md 6.1 KB
- workers-logs.md 5.5 KB
- workers-rpc.md 5.5 KB
- workers-static-assets.md 4.8 KB
- workers.md 8.6 KB
- workflows.md 6.5 KB
- wrangler.md 7.8 KB
- zaraz.md 5.0 KB
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.
- 12d ago First seen · 136 lines · 262 tokens per session scan A afcbdfd5cec2
stack-cloudflare is a skill published in the GitHub repository e-t-y-b/etyb-skills (10 stars, last pushed 2mo ago), licensed MIT. It adds 262 tokens to every session and 3,965 once invoked, about $0.0013 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 2026-08-31.
Other skills, from other repositories
architecture-paradigm-microservices
Applies microservices for independent deployment and per-service scaling. Use when teams need autonomous release cycles with distinct capability scaling needs.
architecture-paradigm-serverless
Applies serverless FaaS patterns for event-driven workloads. Use when designing bursty workloads with minimal infrastructure and pay-per-execution cost model.
configure-nginx
Configure Nginx as a web server and reverse proxy. Covers static file serving, reverse proxy to upstream services, SSL/TLS termination with Let's Encrypt, location blocks, load balancing, rate limiting, and security headers. Use when serving static files in production, reverse proxying to backend services (Node.js…
configure-reverse-proxy
Configure reverse proxy patterns across multiple tools including Nginx, Traefik, and ShinyProxy. Covers WebSocket proxying, path-based and host-based routing, SSL termination, and Docker label auto-discovery. Use when routing multiple services behind a single entry point, proxying WebSocket connections (Shiny…
terraform-infrastructure
Structures, writes, and reviews Terraform infrastructure code. Covers module layout, remote state, workspace strategy, variable and secrets handling, CI plan/apply pipeline, naming conventions, and multi-region deployment patterns (provider aliases, per-region state, failover strategies), while delegating shared risk…
configure-api-gateway
Deploy and configure an API gateway (Kong or Traefik) to handle API traffic management, authentication, rate limiting, request/response transformation, and routing. Covers plugin configuration, upstream services, consumer management, and integration with existing infrastructure. Use when multiple backend services need…