# I built an open-source, offline AST guardrail & MCP server (Aveyx v3.0.1) to stop AI coding agents from faking tests, leaking secrets, and breaking prod

> Source: <https://dev.to/shaikhamanulla/i-built-an-open-source-offline-ast-guardrail-mcp-server-aveyx-v301-to-stop-ai-coding-agents-4270>
> Published: 2026-11-21 05:25:46+00:00

**Key takeaway:** Aveyx v3.0.1 wraps Babel AST checks into an offline MCP server with 10 tools, letting coding agents mechanically block test-skipping, secret leaks and protected-config edits instead of relying on prompt rules.

If you use Claude Code, Cursor, Windsurf, or Copilot on a multi-file repo, you've probably seen them:

`it.skip` / `|| exit 0` to make tests "pass"
I just shipped **Aveyx v3.0.1** (MIT, Node 20+ ESM, 100% offline, only 3 `@babel/*` runtime deps) to catch these mechanically instead of relying on prompt rules alone:

`aveyx impact <file>` — Computes transitive reverse dependencies (ESM, CJS, re-exports, dynamic `import()`) and lists every affected test file.` aveyx compact` — Strips function bodies via AST while preserving signatures, types, and PageRank callers.`aveyx review --strict` — Babel AST anti-tamper + secret scanner + protected-config guard (outputs OASIS SARIF 2.1.0).`aveyx audit-bugs` & `aveyx scan-secrets` — AST SQLi / command injection / React hook leak + credential scanner.`aveyx checkpoint` / `rollback` — Atomic snapshots (including `.env` and git hooks) before risky refactors.`aveyx install-hooks --pre-push` — Blocks bad commits/pushes locally.`aveyx-mcp` — Native stdio MCP server with 10 tools and 44 prompts.
Quick start:

```
bash
npx aveyx init
npx aveyx doctor
![ ](https://dev-to-uploads.s3.us-east-2.amazonaws.com/uploads/articles/t7z6za4gca4zso68bjj7.png)
```


