Atlas / Skills / jeremylongshore / Juicebox Rate Limits

Juicebox Rate LimitsSAFE

skills/jeremylongshore/juicebox-rate-limits

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
1.16.0
Hosts
1 documented
License
MIT
Stars
2,824
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 · 2026-10-09
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: juicebox-rate-limits
description: 'Implement Juicebox rate limiting.

  Trigger: "juicebox rate limit", "juicebox 429", "juicebox throttle".

  '
allowed-tools: Read, Write, Edit
version: 1.16.0
license: MIT
author: Jeremy Longshore <[email protected]>
tags:
- saas
- recruiting
- juicebox
compatibility: Designed for Claude Code
---
# Juicebox Rate Limits

## Overview

Juicebox's AI-powered data analysis API enforces plan-tiered rate limits across dataset uploads, analysis triggers, and result retrieval. Heavy analytical workloads like running comparative analyses across multiple datasets or batch-processing survey results hit the analysis trigger limit first. The enrichment endpoints for augmenting datasets with external data sources have separate, lower caps, making it essential to prioritize enrichment calls and batch analysis runs during off-peak windows.

## Rate Limit Reference

| Endpoint | Limit | Window | Scope |
|----------|-------|--------|-------|
| Dataset upload | 20 req | 1 minute | Per API key |
| Analysis trigger | 30 req | 1 minute | Per API key |
| Result retrieval | 120 req | 1 minute | Per API key |
| Data enrichment | 15 req | 1 minute | Per API key |
| Export download | 10 req | 1 minute | Per API key |

## Rate Limiter Implementation

```typescript
class JuiceboxRateLimiter {
  private tokens: number;
  private lastRefill: number;
  private readonly max: number;
  private readonly refillRate: number;
  private queue: Array<{ resolve: () => void }> = [];

  constructor(maxPerMinute: number) {
    this.max = maxPerMinute;
    this.tokens = maxPerMinute;
    this.lastRefill = Date.now();
    this.refillRate = maxPerMinute / 60_000;
  }

  async acquire(): Promise<void> {
    this.refill();
    if (this.tokens >= 1) { this.tokens -= 1; return; }
    return new Promise(resolve => this.queue.push({ resolve }));
  }

  private refill() {
    const now = Date.now();
    this.tokens = Math.min(this.max, this.tokens + (now - this.lastRefill) * this.refillRate);
    this.lastRefill = now;
    while (this.tokens >= 1 && this.queue.length) {
      this.tokens -= 1;
      this.queue.shift()!.resolve();
    }
  }
}

const analysisLimiter = new JuiceboxRateLimiter(25);
const enrichLimiter = new JuiceboxRateLimiter(12);
```

## Retry Strategy

```typescript
async function juiceboxRetry<T>(
  limiter: JuiceboxRateLimiter, fn: () => Promise<Response>, maxRetries = 3
): Promise<T> {
  for (let attempt = 0; attempt <= maxRetries; attempt++) {
    await limiter.acquire();
    const res = await fn();
    if (res.ok) return res.json();
    if (res.status === 429) {
      const retryAfter = parseInt(res.headers.get("Retry-After") || "15", 10);
      const jitter = Math.random() * 3000;
      await new Promise(r => setTimeout(r, retryAfter * 1000 + jitter));
      continue;
    }
    if (res.status >= 500 && attempt < maxRetries) {
      await new Promise(r => setTimeout(r, Math.pow(2, attempt) * 2000));
      continue;
    }
    throw new Error(`Juicebox API ${res.status}: ${await res.text()}`);
  }
  throw new Error("Max retries exceeded");
}
```

## Batch Processing

```typescript
async function batchAnalyzeDatasets(datasetIds: string[], query: string, batchSize = 5) {
  const results: any[] = [];
  for (let i = 0; i < datasetIds.length; i += batchSize) {
    const batch = datasetIds.slice(i, i + batchSize);
    const batchResults = await Promise.all(
      batch.map(id => juiceboxRetry(analysisLimiter, () =>
        fetch(`${BASE}/api/v1/datasets/${id}/analyze`, {
          method: "POST", headers,
          body: JSON.stringify({ query }),
        })
      ))
    );
    results.push(...batchResults);
    if (i + batchSize < datasetIds.length) await new Promise(r => setTimeout(r, 8000));
  }
  return results;
}
```

## Error Handling

| Issue | Cause | Fix |
|-------|-------|-----|
| 429 on analysis trigger | Exceeded 30 req/min analysis cap | Queue analyses, space 3s apart |
| 429 on enrichment | Enrichment limit (15/min) is lowest | Batch enrichments separately with wider spacing |
| Upload timeout | Dataset exceeds 50MB | Compress CSV, use chunked upload endpoint |
| Analysis still processing | Complex query on large dataset | Poll status every 10s, timeout at 10 min |
| 403 on export | Plan does not include export feature | Verify plan tier supports data export |

## Prerequisites

- An approved sandbox workload, synthetic records, current quota limits, source/destination allowlists, suppression controls, and a named operator for pause and rollback.

## Instructions

1. Exercise rate-limit behavior only with synthetic fixtures and bounded request budgets; reject real-record export or unapproved destinations.
2. Apply idempotency keys, exponential backoff, and aggregate-only telemetry; verify suppression and `contacts_exported=0` at each probe.
3. Stop the canary on unexpected quota, scope, policy, or retention drift, then cancel queued work and revoke temporary access.
4. Retain only redacted aggregate evidence and delete sandbox artifacts after the approved test window.

## Output

Produce a rate-limit receipt with environment, request budget, aggregate response/error counts, suppression and no-export assertions, pause/rollback action, owner approval, and retention/deletion proof. Exclude queries, contacts, and credentials.

## Examples

`env=ci-synthetic; budget=30rpm; retries=backoff; suppression=pass; contacts_exported=0; queued_jobs=cancelled; cleanup=verified` is an acceptable receipt.

## Resources

- Juicebox API Documentation

## Next Steps

See `juicebox-performance-tuning`.
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 (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.

Audited 2026-10-09 · audit v0.4.1 · source sha 4f83675ca38afull audit observations/trust-audit/skill/jeremylongshore__juicebox-rate-limits.json · Report an issue / request a re-scan
05

Audit history

Every audit this skill has had.

DateSourceVerdictGradeScoreChange
2026-10-094f83675ca38aSAFEB89first audit
06

Questions

What does the Juicebox Rate Limits 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 Juicebox Rate Limits 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 Juicebox Rate Limits access on my machine?

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

Which assistants does Juicebox Rate Limits 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 2026-10-09. The repository is watched, and a new audit runs when it changes — this is the first audit.

Advertisement