Clerk Cost TuningSAFE
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
Overview
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
4f83675ca38aOBSERVED · 2026-10-08Host compatibility
What the documentation claims. We have not run a compatibility test.
| Host | Status | Notes |
|---|---|---|
| claude-code | mentioned |
What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
--- name: clerk-cost-tuning description: 'Optimize Clerk costs and understand pricing. Use when planning budget, reducing costs, or understanding Clerk pricing model. Trigger with phrases like "clerk cost", "clerk pricing", "reduce clerk cost", "clerk billing", "clerk budget". ' allowed-tools: Read, Write, Edit, Grep version: 1.15.0 license: MIT author: Jeremy Longshore <[email protected]> tags: - saas - clerk - cost-optimization compatibility: Designed for Claude Code --- # Clerk Cost Tuning ## Overview Understand Clerk pricing and optimize costs. Clerk charges by Monthly Retained Users (MRU) — a user is counted as retained only when they return 24+ hours after signing up. Covers pricing tiers, MRU reduction strategies, caching to reduce API calls, and usage monitoring. ## Prerequisites - Clerk account active - Understanding of MRU (Monthly Retained Users) - Application usage patterns known ## Instructions ### Step 1: Understand Clerk Pricing Model | Plan | Price | MRU Included | Extra MRU | |------|-------|-------------|-----------| | Free (Hobby) | $0/mo | 50,000 MRU | N/A | | Pro | $25/mo ($20/mo billed annually) | 50,000 MRU | $0.02/MRU | | Business | $300/mo ($250/mo billed annually) | 50,000 MRU | $0.02/MRU | | Enterprise | Custom | Custom | Custom | Key pricing concepts: - **MRU** = unique user who returns to your app 24+ hours after signing up ("first day free" — sign-up-day activity never counts) - Users who sign up and never return are not billed - Users who only visit public pages are not counted - Bot/crawler sessions are not counted - Test/development instances are free and unlimited ### Step 2: Reduce MRU Count ```typescript // Strategy 1: Defer authentication — don't force sign-in until necessary // middleware.ts import { clerkMiddleware, createRouteMatcher } from '@clerk/nextjs/server' const requiresAuth = createRouteMatcher([ '/dashboard(.*)', '/settings(.*)', '/api/protected(.*)', ]) export default clerkMiddleware(async (auth, req) => { // Only require auth for specific routes (not entire site) if (requiresAuth(req)) { await auth.protect() } }) ``` ```typescript // Strategy 2: Use anonymous access for read-only features // app/blog/[slug]/page.tsx import { auth } from '@clerk/nextjs/server' export default async function BlogPost({ params }: { params: { slug: string } }) { const { userId } = await auth() // Check but don't require const post = await db.post.findUnique({ where: { slug: params.slug } }) return ( <article> <h1>{post?.title}</h1> <div>{post?.content}</div> {userId ? <CommentForm /> : <p>Sign in to comment</p>} </article> ) } ``` ### Step 3: Cache to Reduce API Calls ```typescript // lib/user-cache.ts import { cache } from 'react' import { currentUser } from '@clerk/nextjs/server' // Deduplicate within single request (free) export const getUser = cache(async () => { return currentUser() }) // Cross-request caching reduces Backend API calls import { unstable_cache } from 'next/cache' import { clerkClient } from '@clerk/nextjs/server' export const getUserMetadata = unstable_cache( async (userId: string) => { const client = await clerkClient() const user = await client.users.getUser(userId) return user.publicMetadata }, ['user-metadata'], { revalidate: 600 } // 10-minute cache ) ``` ### Step 4: Monitor Usage ```typescript // app/api/admin/clerk-usage/route.ts import { auth, clerkClient } from '@clerk/nextjs/server' export async function GET() { const { has } = await auth() if (!has({ role: 'org:admin' })) { return Response.json({ error: 'Admin only' }, { status: 403 }) } const client = await clerkClient() const users = await client.users.getUserList({ limit: 1 }) return Response.json({ totalUsers: users.totalCount, // Estimate MRU based on recent sign-ins estimatedMRU: 'Check Clerk Dashboard > Billing for actual MRU', dashboardUrl: 'last-active?after=30d', }) } ``` ### Step 5: Clean Up Inactive Users ```typescript // scripts/cleanup-inactive-users.ts import { createClerkClient } from '@clerk/backend' const clerk = createClerkClient({ secretKey: process.env.CLERK_SECRET_KEY! }) async function findInactiveUsers(daysInactive = 90) { const cutoff = Date.now() - daysInactive * 24 * 60 * 60 * 1000 const allUsers = await clerk.users.getUserList({ limit: 500 }) const inactive = allUsers.data.filter( (user) => (user.lastSignInAt || 0) < cutoff ) console.log(`Found ${inactive.length} users inactive for ${daysInactive}+ days`) console.log('Consider: notification campaign, data export, or account cleanup') return inactive } findInactiveUsers() ``` ## Output - Pricing model understood with MRU thresholds - Route-level auth to minimize unnecessary MRU counts - Request-level and cross-request caching reducing API calls - Usage monitoring endpoint for admins - Inactive user identification script ## Error Handling | Issue | Cause | Solution | |-------|-------|----------| | Unexpected bill increase | MRU spike from bot traffic | Add bot detection, restrict auth to needed routes | | Feature limitations | Free tier limits (no SSO, etc.) | Upgrade to Pro ($25/mo, or $20/mo billed annually) | | High API call volume | No caching | Add React `cache()` + `unstable_cache()` | | MRU count mismatch | Counting test users | Use separate dev instance (free, unlimited) | ## Examples ### Cost Estimation Script ```typescript function estimateMonthlyCost(mru: number): string { if (mru <= 50_000) return 'Free tier ($0/mo)' const overage = mru - 50_000 const cost = 25 + overage * 0.02 return `Pro tier: $${cost.toFixed(2)}/mo (${overage.toLocaleString()} extra MRU at $0.02 each)` } console.log(estimateMonthlyCost(15_000)) // "Free tier ($0/mo)" console.log(estimateMonthlyCost(60_000)) // "Pro tier: $225.00/mo (10,000 extra MRU at $0.02 each)" ``` ## Resources - [Clerk Pricing](https://clerk.com/pricin
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
4f83675ca38afull audit observations/trust-audit/skill/jeremylongshore__clerk-cost-tuning.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 4f83675ca38a | SAFE | B | 89 | first audit |
Questions
What does the Clerk Cost Tuning skill do?
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
Is Clerk Cost Tuning safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Clerk Cost Tuning access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
Which assistants does Clerk Cost Tuning work with?
Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (4f83675ca38a), read on 2026-10-08. The repository is watched, and a new audit runs when it changes — this is the first audit.