Real, runnable Flue agents on Cloudflare. Every snippet deploys a real ephemeral Worker, exercises it against live services, then tears it down. No mocks.

Demo site: flue.coey.dev, a guided tour of Flue on Cloudflare, auto-generated from this repo.

examples/   one CF product per folder, smallest Flue agent that uses it
recipes/    compositions, Flue + multiple primitives + receipts/proofs
templates/  forkable starters, production-shape, fork-and-ship

git clone https://github.com/acoyfellow/flue-snippets
cd flue-snippets
bun install         # or: npm install

# https://dash.cloudflare.com/profile/api-tokens, Workers Scripts:Edit + Workers AI:Read
export CLOUDFLARE_API_TOKEN=...
export CLOUDFLARE_ACCOUNT_ID=...

bun ex:workers-ai   # or: npm run ex:workers-ai

~60 seconds: deploy, run, assert, destroy. ~$0.0001 in Workers AI usage.

The agent it just ran:

// examples/workers-ai/workers-ai.ts
import type { FlueContext } from '@flue/sdk/client';

export const triggers = { webhook: true };

export default async function ({ payload, env }: FlueContext) {
  const out = await env.AI.run('@cf/moonshotai/kimi-k2.6', {
    prompt: payload.message ?? 'Say hi.',
  });
  return { answer: out.response };
}

One Flue agent, one Cloudflare binding.

Flue + a Cloudflare primitive + an open-source receipt/proof layer (@acoyfellow/lab, gateproof).

Each recipe's README explains what it composes, what it proves, how to run it.

bun ex:<name>    # examples/<name>/run-e2e.sh
bun rx:<name>    # recipes/<name>/run-e2e.sh
bun tpl:<name>   # templates/<name>/run-e2e.sh

See package.json for the full list.

bun run update

That repeatable maintenance command runs bun update in the repo and in site/, refreshes the homepage's visible “Last dependency refresh” timestamp when Bun bumps package manifest versions, then runs the site production build. If the Flue CLI/runtime or a dependency used by a snippet changes shape, also update the affected example/recipe/template and run its live E2E (bun ex:<name>, bun rx:<name>, or bun tpl:<name>) before shipping.

Every run-e2e.sh does the same five things:

1. flue build --target cloudflare, emits the Worker entrypoint + per-agent DO classes.
2. alchemy deploy, declares the Worker, bindings, and vars; bundles; prints the URL.
3. Warmup, polls /health, then POSTs /agents/<name>/warmup with retries (absorbs route propagation + Workers AI cold start).
4. Assert, examples curl-and-grep; recipes run a gateproof.plan.ts with a probe.ts (pure fetch + JSON).
5. alchemy destroy, tears the Worker, bindings, and state down. Trapped on EXIT INT TERM.

Wrangler is not invoked. alchemy owns the resource graph; Flue emits the entrypoint module that alchemy bundles.

.github/workflows/e2e.yml is workflow_dispatch-only. From the Actions tab, pick all, examples, recipes, templates, or a single target. max-parallel: 1 because Workers AI rate-limits hard under parallel load on a personal account.

Secrets:

• CF_API_TOKEN_E2E, Workers Scripts:Edit + Workers AI:Read, plus permissions for any product-specific targets you enable (R2 / D1 / KV / Queues / AI Gateway / Vectorize / Browser Rendering / Worker Loader / Hyperdrive / Email).
• CF_ACCOUNT_ID_E2E, Cloudflare account ID.
• EMAIL_FROM, EMAIL_TO, only needed if you enable examples/email-workers. Without them, the send call returns a structured error and the assertion still passes (it accepts either a real send or a structured E_* code).

Does it really deploy? Yes. Each run-e2e.sh calls alchemy deploy, hits a real *.workers.dev URL, then alchemy destroys it. CI does the same. There is no mock layer.

What does it cost? ~$0.0001 per snippet per run (Workers AI llama-scout call). Free tier is plenty for the entire matrix.

Why no wrangler? alchemy is the resource graph (Workers + bindings + vars, declared in TypeScript, with destroy). Flue is the agent runtime (emits the Worker entrypoint module). The two compose cleanly; wrangler would duplicate alchemy's job.

Why does CI run sequentially? Workers AI rate-limits aggressively on personal accounts under parallel load. max-parallel: 1 keeps the matrix green.

Can I run only one? Yes, every example, recipe, and template is independent. bun ex:<name>, bun rx:<name>, or bun tpl:<name>. Or trigger a single target from the Actions dropdown.

See CONTRIBUTING.md. Security issues: SECURITY.md.

MIT.