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.
git clone --depth 1 https://github.com/louisbrulenaudet/monorepo-templateWrote 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/rules/louisbrulenaudet/monorepo-template/previews)<a href="https://agentmods.dev/rules/louisbrulenaudet/monorepo-template/previews"><img src="https://agentmods.dev/badge/rules/louisbrulenaudet/monorepo-template/previews/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/rules/louisbrulenaudet/monorepo-template/previews"><img src="https://agentmods.dev/badge/rules/louisbrulenaudet/monorepo-template/previews.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 | $0.00000 | $0.01589 |
| Opus 5.5 | $0.00000 | $0.00636 |
| Sonnet 5.5 | $0.00000 | $0.00318 |
| Haiku 4.5 | $0.00000 | $0.00159 |
Grade A, and why
previews 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 today.
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 — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Worker Previews
Branch and PR testing uses Worker Previews: an isolated Preview under the production Worker (<app>-production), with its own bindings, URL (<preview>-<app>-production.<subdomain>.workers.dev), and observability, built from each app's ctx.isPreview branch (workers-config.mdc, Preview branch) and deployed with cf previews deploy. The pipeline is .github/workflows/preview.yml, the scripts .github/actions/previews/; read them for the steps. Never use Version URLs (cf workers versions create --preview-alias) or an extra mode for branch testing: both run on shared resources.
Build, deploy, delete
- Every
cfPreview command that readscloudflare.config.tscarries--mode production(cf build,cf previews deploy).cf previews deleteand the Wrangler-onlypreview secret putread no config and name the Worker instead (--worker/--worker-name <app>-production). deploy-previews.shwalks the apps inmonorepo.deployOrder. Each app is built without credentials -env -u CLOUDFLARE_API_TOKEN -u CLOUDFLARE_ACCOUNT_ID -u CF_ACCESS_CLIENT_ID -u CF_ACCESS_CLIENT_SECRET CLOUDFLARE_PREVIEW_BUILD=true pnpm --filter=<app> exec cf build --mode production, plus theVITE_*values for a frontend - then deployed with credentials:pnpm --filter=<app> exec cf previews deploy <name> --prebuilt --mode production.CLOUDFLARE_PREVIEW_BUILD=trueis cf's internal, undocumented variable that makes a build a Preview build (ctx.isPreview, recorded asbuildContext.isPreview).cf previews deploy --prebuiltrefuses Build Output that is not a Preview build, so if cf drops the variable the deploy fails loudly instead of shipping production values to a Preview; converselycf deploy,cf workers versions create, andcf workers triggers deployrefuse a Preview build.cf previews deployprints one JSON object on stdout (type: "preview",preview_urls,deployment_id, …) and build logs on stderr; the URL is.preview_urls[0]. It takes no--message.- The SPA's
VITE_API_BASE_URLis the gateway Preview URL, which is why the gateway is previewed first and the frontend built afterwards. - One smoke, after the last deploy:
deploy-previews.shhands every Preview URL to.github/actions/lib/smoke.mjs, the same smoke as CD and skillrun-app, so the gateway check also proves its CORS allowlist admits the SPA Preview origin. A--urltarget is retried for ~25 s on an error, a 404, or a 52x while the new hostname starts serving. A broken gateway therefore costs one SPA build before the run fails. - The PR comment is
changesets/action/pr-comment(update-id: worker-previews), fed by the deploy step'sbodyoutput; locally the same URL table goes to stdout. It searches only the first 30 comments (quality-report.mdc). delete-previews.shrunspnpm exec cf previews delete <preview> --worker <app>-production --forcefrom the repo root for every app, attempting each even when one fails, with no config file.--forceis required: without a TTY, cf exits 1 on an unconfirmed destructive command. Locally it authenticates through cf (pnpm run login, or exportedCLOUDFLARE_API_TOKEN/CLOUDFLARE_ACCOUNT_ID).
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.
- today Changed f70dd332ba6b
- 13d ago Changed · +2 lines 6d7d6aedcf20
- 16d ago Changed · -5 lines · -1,528 tokens per session 314c0a4fb1f8
- 17d ago Changed · -12 tokens per session be89dc7493ea
- 18d ago Changed · +125 tokens per session 0556ba7101c1
- 20d ago First seen · 43 lines · 1,415 tokens per session scan A b4e200e84ef9
previews is a cursor rule published in the GitHub repository louisbrulenaudet/monorepo-template (19 stars, last pushed today), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,589 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 2026-09-26.
Other cursor rules, from other repositories
cloudflare-workers-auto
Cloudflare Workers development standards and best practices.
cloudflare-workers-hono-auto
Cloudflare Workers with Hono framework development standards and best practices.
ankra-cli
Ankra CLI rules and best practices for managing Kubernetes clusters via the Ankra platform.
cloudflare-email-telegram-cursorrules-prompt-file
Cursor rules for setting up email-to-Telegram forwarding via Cloudflare Email Routing and Workers using the mail2tg CLI.
paperfit
PaperFit project rule for LaTeX visual typesetting optimization.
swift-argument-parser
Documentation and usage patterns for Swift Argument Parser for command-line tools.