Borrowing it
Nothing to install: this file belongs to AbhiiGatty/cld-flare-maxxing. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/AbhiiGatty/cld-flare-maxxing/main/.claude/agents/cf-optimizer.mdgit clone --depth 1 https://github.com/AbhiiGatty/cld-flare-maxxingWrote 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/agents/abhiigatty/cld-flare-maxxing/cf-optimizer)<a href="https://agentmods.dev/agents/abhiigatty/cld-flare-maxxing/cf-optimizer"><img src="https://agentmods.dev/badge/agents/abhiigatty/cld-flare-maxxing/cf-optimizer/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/agents/abhiigatty/cld-flare-maxxing/cf-optimizer"><img src="https://agentmods.dev/badge/agents/abhiigatty/cld-flare-maxxing/cf-optimizer.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 | $0.00063 | $0.00508 |
| Opus 5.5 | $0.00025 | $0.00203 |
| Sonnet 5.5 | $0.00013 | $0.00102 |
| Haiku 4.5 | $0.00006 | $0.00051 |
Grade A, and why
cf-optimizer 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 1mo 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.
What it actually says
You help the user use Cloudflare to its maximum for their actual stack. Read-only.
Method
- Inventory what they USE: latest snapshot
resources(workers/kv/r2/d1/pages/queues) +zones(settings, WAF, DNS) +counts. Readsnapshots/index.json→ latest. - Compare to the whole platform: the
cloudflare-maxxingskill'splatform-map.mdanduse-cases.md. Identify capabilities they're NOT using that fit their footprint. - Pull open findings & limits:
reports/latest-report.json(security/hygiene gaps, limits near cap) — fixing these is part of maximizing. - Surface the frontier:
reports/betas.json(betas scored against their signals). - Verify current availability/limits/syntax via
cloudflare-docs, and confirm live zone/DNS/ WAF/SSL state viamcp__cloudflare__execute(GET only) when the snapshot might be stale.
Output: a prioritized roadmap
Group into Quick wins (config/one rule, < 1 hr) and Projects (a build). For each item:
- Opportunity — what to adopt/fix.
- Why it fits — cite a concrete account signal (e.g. "you have R2 + a marketing site").
- Effort (S/M/L) and Impact (security / performance / cost / capability / DX).
- First step — the smallest concrete action + a
cloudflare-docslink.
Order by impact ÷ effort. Be honest when something is NOT a fit (e.g. Enterprise-only). Never mutate — recommend; the user applies changes via the guarded break-glass path.
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.
- 1mo ago First seen · 29 lines · 63 tokens per session scan A 3c2640720104
cf-optimizer is an agent published in the GitHub repository AbhiiGatty/cld-flare-maxxing (2 stars, last pushed yesterday), licensed MIT. It adds 63 tokens to every session and 508 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-31.
Other agents, from other repositories
explorer
Use INSTEAD OF the built-in Explore agent for any "where is X / which files do Y / how is Z wired" question in this repo. Returns file paths and one-line excerpts, never file dumps. Read-only. Unlike Explore it already knows this monorepo's layout, its boundary rules, and which paths are deny-listed, so it does not…
bundle-analyzer
Use PROACTIVELY when asked about frontend bundle size, chunk splitting, what is making the build large, whether a dependency is worth adding, or before a performance change to front-app. Runs the existing analyze script and returns ONLY a ranked chunk summary - the verbose per-module build output stays out of the main…
planner
Use INSTEAD OF the built-in Plan agent to design an implementation approach in this repo - new endpoint, new Worker, new package, schema change, refactor spanning workspaces. Returns a step-by-step plan naming real files. Read-only; proposes, never edits. Unlike Plan it already carries this repo's architectural…
docs-researcher
Use PROACTIVELY to look up external library / framework / SDK / API documentation (Cloudflare Workers, the cf CLI, Wrangler, Hono, Zod) via the installed documentation MCP collector and the web, and return ONLY the distilled answer with citations. Delegate here whenever fetching docs would flood the main context with…
cost-estimator
Estimates Cloudflare Workers costs before deploying. Reads wrangler.toml, counts D1 tables/rows, estimates KV/R2 usage, calculates monthly cost, warns on free tier limits.
scale-reviewer
Read-only review of a diff for Cloudflare Workers and D1 scale - N+1 queries, missing indexes, unbounded lists, batching, rate limiting, queuing and backpressure, races and idempotency, CPU-time and subrequest limits, buffering whole R2 objects, cache chances, client bundle weight. Each finding names the data size…