- Home
- Skills
- DevOps & Cloud
- ShipSafe Deployment Auditor
Works with the AI Agents you already use
ShipSafe Deployment Auditor
A read-only pre-deployment audit for Next.js and JavaScript/TypeScript projects. Find release blockers, risky configuration, and missing checks before shipping to Vercel.
Free
ShipSafe Deployment Auditor
Example session with this skill installed
Perform a pre-deployment audit on my Next.js project to check for Vercel readiness and environment variable risks.
- Read your context and instructions
- Compiled the shipsafe deployment auditor
I have completed the static pre-deployment audit. The project is identified as a Next.js (App Router) application using pnpm.
Deployment Readiness Report: READY WITH RISKS
Executive Summary
The project has valid build scripts and a consistent lockfile. However, there are potential environment variable exposures and missing configuration for remote images that will cause Vercel deployment warnings.
Prioritized Findings
| Severity | Type | Path | Finding | Remediation |
|---|---|---|---|---|
| HIGH | Security | src/lib/api.ts:12 | NEXT_PUBLIC_STRIPE_KEY detected. Public prefix suggests client-side exposure of a sensitive key. | Move to server-only env var or verify if this is the Publishable Key. |
| MEDIUM | Config | next.config.js | Missing images.remotePatterns. External images from images.unsplash.com will fail to load. | Add images.remotePatterns for Unsplash to the config file. |
| LOW | Build | package.json | lint script found but not included in build or CI workflow. | Add pnpm lint to the pre-build step to catch errors early. |
| INFO | Docs | .env.example | 3 variables used in source are missing from the example file. | Update .env.example to include DATABASE_URL and API_SECRET_KEY placeholders. |
Checks Passed
- Lockfile consistency (
pnpm-lock.yaml) - Build script presence
- TypeScript configuration (
tsconfig.jsonexists) - No hardcoded AWS credentials found in tracked source
Pre-deploy Checklist
- Verify
NEXT_PUBLIC_STRIPE_KEYis intended for the browser. - Update
next.config.jswith required image domains. - Sync
.env.examplewith local environment keys.
Next steps
- Propose a fix for the
next.config.jsimage patterns. - Perform a deep dive into
src/lib/api.tsto check how the Stripe key is used. - Generate a diff to sync the
.env.examplefile.
Connects securely to your tools. The creator never sees your data.
About this skill
Deploying to Vercel or production environments often involves a "trial by fire" where build errors or exposed secrets only appear after a failed pipeline. This skill performs a non-invasive, read-only audit of JavaScript and TypeScript projects to identify these blockers before you push code. It prioritizes evidence-based findings over generic linting rules, focusing specifically on configuration risks in Next.js and Vercel environments.
What it does
- Pre-deployment auditing identifies missing build scripts, lockfile mismatches, and configuration errors that cause deployment failures.
- Environment variable hygiene flags risky
NEXT_PUBLIC_prefixes and identifies missing keys in.env.examplewithout reading or exposing actual values. - Next.js configuration review checks for App/Pages router conflicts, image remote patterns, and static export issues.
- Vercel readiness assessment inspects
vercel.jsonfor complex rewrites, redirects, and runtime settings that might trigger production issues. - Safety-first remediation proposes minimal-scope fixes and only modifies files after explicit user approval of a provided diff.
How it works
- Identify stack by inspecting lockfiles,
package.json, and framework configurations to determine the project type. - Execute static preflight using a dependency-free Python script to analyze repository structure and configuration signals.
- Contextual review cross-references script findings with source code to validate risks in build scripts, dependencies, and quality gates.
- Prioritize findings by categorizing issues from
BLOCKERtoINFOwith exact file paths and evidence-based impact statements. - Generate report using a structured template that highlights readiness status and provides a pre-deploy checklist.
Frameworks & tools
Specifically designed for JavaScript and TypeScript projects including Next.js, React, and Vite. It supports pnpm, yarn, npm, and bun package managers and targets Vercel deployment configurations.
Why this beats prompting it yourself
Generic prompts often miss the nuances of Vercel-specific configuration or accidentally leak environment variables during analysis. This skill uses a dedicated static analysis script and a strict safety contract to ensure secrets remain private while providing evidence-backed remediation rather than generic advice.
Use cases
- Audit a Next.js application for release blockers before a major production push.
- Review environment variable exposure risks to ensure no secrets are prefixed for client-side access.
- Resolve package manager and lockfile ambiguities that cause CI/CD pipeline failures.
- Validate Vercel-specific settings like middleware, redirects, and runtime configurations.
Known limitations
Does not perform runtime behavior analysis or live network requests. It cannot verify if a deployment is 100% safe or confirm exploitable vulnerabilities without manual proof. Project must be JS/TS.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for DevOps & Cloud.
Read-only static analysis only. Does not run code, perform network requests, or access live Vercel APIs. Limited to JavaScript and TypeScript projects. Cannot prove runtime exploitability.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 9 days ago
- Free with an Agensi account
- Software
Python
Git