Atlas / Skills / jeremylongshore / Juicebox Multi Env Setup

Juicebox Multi Env SetupSAFE

skills/jeremylongshore/juicebox-multi-env-setup

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-multi-env-setup
description: 'Configure Juicebox multi-environment.

  Trigger: "juicebox environments", "juicebox staging".

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

## Overview

Juicebox AI analysis requires environment separation to enforce workspace isolation, control data access, and prevent accidental exports of sensitive datasets. Development works with sample datasets and strict result limits for fast iteration, staging connects to full production data but disables export functionality, and production enables all features with full export capabilities. Each environment uses isolated workspaces so analysis experiments in dev never affect production workspace state or user-facing reports.

## Environment Configuration

```typescript
const juiceboxConfig = (env: string) => ({
  development: {
    apiKey: process.env.JB_KEY_DEV!, baseUrl: "https://api.dev.juicebox.work/v1",
    workspaceId: process.env.JB_WORKSPACE_DEV!, resultLimit: 5, exportEnabled: false,
  },
  staging: {
    apiKey: process.env.JB_KEY_STG!, baseUrl: "https://api.staging.juicebox.work/v1",
    workspaceId: process.env.JB_WORKSPACE_STG!, resultLimit: 20, exportEnabled: false,
  },
  production: {
    apiKey: process.env.JB_KEY_PROD!, baseUrl: "https://api.juicebox.work/v1",
    workspaceId: process.env.JB_WORKSPACE_PROD!, resultLimit: 50, exportEnabled: true,
  },
}[env]);
```

## Environment Files

```text
# Per-env files: .env.development, .env.staging, .env.production
JB_KEY_{DEV|STG|PROD}=<api-key>
JB_WORKSPACE_{DEV|STG|PROD}=<workspace-id>
JB_BASE_URL=https://api.{dev.|staging.|""}juicebox.work/v1
JB_EXPORT_ENABLED={false|false|true}
JB_DATASET_SCOPE={sample|full|full}
```

## Environment Validation

```typescript
function validateJuiceboxEnv(env: string): void {
  const suffix = { development: "_DEV", staging: "_STG", production: "_PROD" }[env];
  const required = [`JB_KEY${suffix}`, `JB_WORKSPACE${suffix}`, "JB_BASE_URL"];
  const missing = required.filter((k) => !process.env[k]);
  if (missing.length) throw new Error(`Missing Juicebox vars for ${env}: ${missing.join(", ")}`);
}
```

## Promotion Workflow

```bash
# 1. Run analysis queries against sample data in dev
curl -X POST "$JB_BASE_URL/analyze" \
  -H "Authorization: Bearer $JB_KEY_DEV" -d @test-query.json

# 2. Validate same queries against full dataset in staging (exports blocked)
curl -X POST "$JB_BASE_URL/analyze" \
  -H "Authorization: Bearer $JB_KEY_STG" -d @test-query.json | jq '.resultCount'

# 3. Verify workspace isolation — staging results don't appear in prod
curl "$JB_BASE_URL/workspaces/$JB_WORKSPACE_PROD/reports" \
  -H "Authorization: Bearer $JB_KEY_PROD" | jq 'length'

# 4. Deploy to production with export capabilities enabled
JB_EXPORT_ENABLED=true npm run deploy -- --env production
```

## Environment Matrix

| Setting | Dev | Staging | Prod |
|---------|-----|---------|------|
| Dataset | Sample (100 records) | Full dataset | Full dataset |
| Result Limit | 5 | 20 | 50 |
| Export Enabled | No | No | Yes |
| Workspace | Isolated dev | Isolated staging | Production |
| API Tier | Free | Standard | Enterprise |
| Report Sharing | Disabled | Internal only | Full access |

## Error Handling

| Issue | Cause | Fix |
|-------|-------|-----|
| 403 on export endpoint | Export disabled in non-prod env | Expected; exports only work in production |
| Workspace not found | Workspace ID mismatch for env | Verify `JB_WORKSPACE_*` matches Juicebox admin panel |
| Result limit exceeded | Query returns more than env cap | Reduce query scope or increase limit in config |
| Stale sample data in dev | Dev dataset not refreshed | Run `npm run seed:dev` to reload sample data |
| API key scope error | Key generated for wrong workspace | Regenerate API key in correct workspace settings |

## Prerequisites

- Approved environment boundaries, sandbox-only synthetic fixtures, secret references rather than literal keys, source/destination allowlists, suppression controls, and a tested rollback revision.

## Instructions

1. Keep development and staging datasets synthetic; reject cross-environment credentials, sources, destinations, and real-record exports.
2. Run deployment probes with aggregate metrics only and confirm suppression, data minimization, and `contacts_exported=0` before promotion.
3. Release through one sandbox canary, monitor quota and policy drift, and stop immediately on scope, retention, or authorization failures.
4. Promote only after owner approval; restore the prior revision, revoke temporary access, and delete staged fixtures if validation fails.

## Output

Produce an environment receipt with environment-to-workspace mapping, fixture type, source/destination approval, suppression and no-export assertions, canary outcome, approver, retention/deletion proof, and rollback reference. Exclude keys and contact-level data.

## Examples

`env=staging; workspace=isolated-synthetic; source=approved; destination=none; suppression=pass; contacts_exported=0; rollback=release-r31` is safe to promote for approval.

## Resources

- [Juicebox Docs](https://docs.juicebox.work)
- Juicebox API Reference

## Next Steps

See `juicebox-deploy-integration`.
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 (1)

INFOPrompt injection · prompt.authority_framing · CWE-94, CWE-1427
SKILL.md:93
| Report Sharing | Disabled | Internal only | Full access |

Gates applied: no_behavioural_pass.

Audited 2026-10-09 · audit v0.4.1 · source sha 4f83675ca38afull audit observations/trust-audit/skill/jeremylongshore__juicebox-multi-env-setup.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 Multi Env Setup 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 Multi Env Setup 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 Multi Env Setup access on my machine?

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

Which assistants does Juicebox Multi Env Setup 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