skinny.cloud — Lean AI ops
// lean ai opsYour AI is running fat.<br>We built the cure — because we got the bill first.<br>zsh — claude-diet
.arrow]:transition-transform [&>.arrow]:duration-[260ms] [&>.arrow]:[transition-timing-function:cubic-bezier(0.22,1,0.36,1)] hover:[&>.arrow]:translate-x-1 bg-signal text-ink-950 hover:bg-signal-dim hover:text-ink-50 animate-[signal-pulse_2.4s_ease-out_2s_3] text-base px-7 py-3.5" href="/diet">Run the audit →▮.arrow]:transition-transform [&>.arrow]:duration-[260ms] [&>.arrow]:[transition-timing-function:cubic-bezier(0.22,1,0.36,1)] hover:[&>.arrow]:translate-x-1 bg-transparent text-ink-50 border border-ink-600 hover:border-ink-400 text-base px-7 py-3.5" href="/about">Read the receipts
Big Tech wants you to buy more context. We want you to need less.
// how it works<br>01 / run it<br>One command. No install. No account.<br>Paste it in your terminal. The script fetches, runs, and tells you exactly what your Claude Code setup loads before you type a prompt.
02 / diet card<br>Your runtime diet card prints immediately<br>No dashboard. No account. The card prints right in the terminal — largest surfaces, hidden memory, hooks, skills, and the top suspect identified.
03 / rules + report<br>Rules generated. Report saved. Done.<br>Three next-session rules written from your actual setup. A full report folder dropped in your project directory — card, worksheet, baseline, passport.
diet — audit
$ curl -sL https://skinny.cloud/diet.sh | bash
skinny.cloud Runtime Diet Autopilot<br>Fetching scripts from github.com/<br>skinny-cloud/runtime-diet-autopilot... ◐
[1/4] Baseline generated.<br>[2/4] Extracting top surfaces and<br>hidden memory...<br>[3/4] Writing worksheet, card,<br>passport, rule, index...
diet — audit
$ curl -sL https://skinny.cloud/diet.sh | bash
skinny.cloud Runtime Diet Autopilot<br>Fetching scripts from github.com/<br>skinny-cloud/runtime-diet-autopilot... ◐
[1/4] Baseline generated.<br>[2/4] Extracting top surfaces and<br>hidden memory...<br>[3/4] Writing worksheet, card,<br>passport, rule, index...
diet — card
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓┃ My Claude Runtime Diet Card ┃┣━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┫┃ Largest visible surface ┃┃ ~/.claude/projects - 819MB ┃┃ ┃┃ Hidden project memory total biggest invisible cost ┃┃ 818.2MiB across 24 projects ┃┃ ┃┃ Largest hidden project memory ┃┃ -home-axw-ASIF - 256MB (381 files) ┃┃ ┃┃ Hooks: 2 | MCP servers: 2 | Skills: 35 | Agents: 2 ┃┃ ┃┃ Top suspect ┃┃ Long-lived session transcript; old branches make the next session ┃┃ heavier than the prompt looks. ┃┗━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┛
diet — lean
[4/4] Done.
Runtime Diet Report written to:<br>~/projects/dx3/<br>runtime-diet-report-20260508
Open first:<br>my-runtime-diet-card.md<br>NEXT-SESSION-RULE.md
Next session rules (generated):<br>1) Start fresh, don't drag old branches<br>2) Prune ~/.claude/projects — 818MB<br>3) Cap MCP + skills + agents per task
Layer 1<br>Runtime Diet Autopilot<br>Audit your Claude Code context in 60 seconds. Free.<br>One shell script scans your CLAUDE.md files, hooks, and memory surfaces, then hands you a Runtime Diet Card — exactly where your tokens are going and what to cut first. No account. No email. No catch.<br>.arrow]:transition-transform [&>.arrow]:duration-[260ms] [&>.arrow]:[transition-timing-function:cubic-bezier(0.22,1,0.36,1)] hover:[&>.arrow]:translate-x-1 bg-transparent text-signal border border-signal hover:bg-signal hover:text-ink-950 text-sm px-5 py-2.5" href="/diet">Get the script →
curlwgetinspect<br>Copy<br>$ curl -sL https://skinny.cloud/diet.sh | bash
$0/mo saved<br>receipts
0 projects<br>in production
0s audit<br>no signup
Across 15 active AI projects, context discipline alone cut Claude costs by $750/month.