Atlas / Skills / jeremylongshore / Customerio Performance Tuning

Customerio Performance TuningSAFE

skills/jeremylongshore/customerio-performance-tuning

Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.

Verdict
SAFE
Grade
B
Trust score
89 /100
Version
1.14.0
Hosts
1 documented
License
MIT
Stars
2,823
01

Overview

Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.

Read from source at commit 4f83675ca38aOBSERVED · 2026-10-08
02

Host compatibility

What the documentation claims. We have not run a compatibility test.

HostStatusNotes
claude-codementioned
03

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: customerio-performance-tuning
description: 'Optimize Customer.io API performance for high throughput.

  Use when improving response times, implementing connection

  pooling, batching, caching, or regional routing.

  Trigger: "customer.io performance", "optimize customer.io",

  "customer.io latency", "customer.io connection pooling".

  '
allowed-tools: Read, Write, Edit, Bash(npm:*), Bash(npx:*), Glob, Grep
version: 1.14.0
license: MIT
author: Jeremy Longshore <[email protected]>
tags:
- saas
- customer-io
- performance
- optimization
compatibility: Designed for Claude Code
---
# Customer.io Performance Tuning

## Output

- A measured delivery/API performance baseline with a scoped optimization, owner, and rollback trigger.
- A capacity decision that preserves consent, message correctness, rate limits, and provider reliability.

## Examples

Measure event throughput, response latency, error rate, and queue depth in a development workspace using synthetic payloads. Change one batch/concurrency parameter, compare against the baseline, and revert if errors, rate limits, or message ordering regress. Do not increase concurrency by replaying production recipient data.

## Overview

Optimize Customer.io API performance for high-volume integrations: HTTP connection pooling, identify deduplication caching, event batching with flush control, fire-and-forget async tracking, and regional routing.

## Prerequisites

- Working Customer.io integration
- Understanding of your traffic patterns and volume
- Monitoring to measure improvement (see `customerio-observability`)

## Performance Targets

| Operation | Baseline | Optimized | Technique |
|-----------|----------|-----------|-----------|
| Single identify | ~200ms | ~80ms | Connection pooling |
| Single track | ~200ms | ~80ms | Connection pooling |
| 100 events batch | ~20s serial | ~500ms | Parallel batching |
| Duplicate identify | ~200ms | ~0ms | Dedup cache |
| Non-critical track | Blocking | Non-blocking | Fire-and-forget |

## Instructions

### Step 1: HTTP Connection Pooling

```typescript
// lib/customerio-pooled.ts
import { TrackClient, RegionUS } from "customerio-node";
import https from "https";

// The customerio-node SDK creates new connections by default.
// Reuse connections with a keep-alive agent.
const agent = new https.Agent({
  keepAlive: true,
  maxSockets: 25,        // Max concurrent connections
  maxFreeSockets: 10,    // Keep idle connections open
  timeout: 30000,        // 30s socket timeout
  keepAliveMsecs: 15000, // TCP keep-alive probe interval
});

// Apply to the SDK by creating a singleton with the agent
// Note: customerio-node doesn't directly accept an agent,
// but we configure Node.js global agent for HTTPS
https.globalAgent = agent;

// Singleton client — one instance = one connection pool
const cio = new TrackClient(
  process.env.CUSTOMERIO_SITE_ID!,
  process.env.CUSTOMERIO_TRACK_API_KEY!,
  { region: RegionUS }
);

export { cio };
```

### Step 2: Identify Deduplication Cache

```typescript
// lib/customerio-dedup.ts
// Skip duplicate identify() calls within a time window

class LRUCache<K, V> {
  private map = new Map<K, V>();
  constructor(private maxSize: number) {}

  get(key: K): V | undefined {
    const val = this.map.get(key);
    if (val !== undefined) {
      // Move to end (most recent)
      this.map.delete(key);
      this.map.set(key, val);
    }
    return val;
  }

  set(key: K, val: V): void {
    this.map.delete(key);
    this.map.set(key, val);
    if (this.map.size > this.maxSize) {
      const oldest = this.map.keys().next().value;
      this.map.delete(oldest!);
    }
  }
}

import { createHash } from "crypto";
import { TrackClient, RegionUS } from "customerio-node";

const identifyCache = new LRUCache<string, number>(10_000);
const DEDUP_TTL_MS = 5 * 60 * 1000;  // 5 minutes

const cio = new TrackClient(
  process.env.CUSTOMERIO_SITE_ID!,
  process.env.CUSTOMERIO_TRACK_API_KEY!,
  { region: RegionUS }
);

export async function dedupIdentify(
  userId: string,
  attrs: Record<string, any>
): Promise<void> {
  // Create a hash of userId + attributes
  const hash = createHash("sha256")
    .update(userId + JSON.stringify(attrs))
    .digest("hex")
    .substring(0, 16);

  const cached = identifyCache.get(hash);
  if (cached && Date.now() - cached < DEDUP_TTL_MS) {
    return; // Skip — identical identify() call within TTL window
  }

  await cio.identify(userId, attrs);
  identifyCache.set(hash, Date.now());
}
```

### Step 3: Batch Processor

```typescript
// lib/customerio-batch.ts
import { TrackClient, RegionUS } from "customerio-node";

interface BatchItem {
  type: "identify" | "track";
  userId: string;
  data: Record<string, any>;
}

export class CioBatchProcessor {
  private buffer: BatchItem[] = [];
  private timer: NodeJS.Timeout | null = null;
  private client: TrackClient;
  private processing = false;

  constructor(
    private readonly maxBatchSize = 100,
    private readonly flushIntervalMs = 3000,
    private readonly concurrency = 15
  ) {
    this.client = new TrackClient(
      process.env.CUSTOMERIO_SITE_ID!,
      process.env.CUSTOMERIO_TRACK_API_KEY!,
      { region: RegionUS }
    );
    this.startFlushTimer();
  }

  add(item: BatchItem): void {
    this.buffer.push(item);
    if (this.buffer.length >= this.maxBatchSize) {
      this.flush();
    }
  }

  async flush(): Promise<void> {
    if (this.processing || this.buffer.length === 0) return;
    this.processing = true;

    const batch = this.buffer.splice(0, this.maxBatchSize);
    const startMs = Date.now();

    // Process in parallel chunks
    for (let i = 0; i < batch.length; i += this.concurrency) {
      const chunk = batch.slice(i, i + this.concurrency);
      const results = await Promise.allSettled(
        chunk.map((item) =>
          item.type === "identify"
            ? this.client.identify(item.userId, item.data)
            : this.client.track(item.userId, item.data)
04

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.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codePASS
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
none-observed
Network
none-observed
Shell
none-observed
Dependencies
pinned
Secrets in source
none-found

Findings (2)

LOWObfuscation / stealth · review.concealment · CWE-506, CWE-94
SKILL.md
.catch((err) => console.error(`CIO async track failed: ${err.message}`))
Why it matters. The fire-and-forget pattern catches tracking errors and only logs them to console, concealing failures from the user with no retry, metric, or surfacing mechanism.
Fix. rewrite it so the instruction says plainly what it does, and asks the user before it acts
LOWObfuscation / stealth · review.concealment · CWE-506, CWE-94
SKILL.md
return; // Skip — identical identify() call within TTL window
Why it matters. The dedup cache silently drops identify calls without any logging or notification, concealing intercepted API calls from the user.
Fix. rewrite it so the instruction says plainly what it does, and asks the user before it acts

Gates applied: no_behavioural_pass.

Audited 2026-10-08 · audit v0.4.1 · source sha 4f83675ca38afull audit observations/trust-audit/skill/jeremylongshore__customerio-performance-tuning.json · Report an issue / request a re-scan
05

Audit history

Every audit this skill has had.

DateSourceVerdictGradeScoreChange
2026-10-084f83675ca38aSAFEB89first audit
06

Questions

What does the Customerio Performance 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 Customerio Performance 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 Customerio Performance Tuning access on my machine?

The audit observed no filesystem, network or shell use at all in its source.

Which assistants does Customerio Performance 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.

Advertisement