inkdust2021

inkdust2021 / opencode-vibeguard

Public

VibeGuard for opencode

19
0
100% credibility
Found Mar 07, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

opencode-vibeguard is a plugin for OpenCode that hides sensitive strings like emails and IDs with placeholders before sending to AI services, while restoring them for local display and tool execution.

How It Works

1
🔍 Discover privacy shield

While using your AI coding helper, you find this tool that keeps your personal info safe from the AI service.

2
📥 Add to your project

Simply place the tool's folder into your work area and tell your AI helper to use it.

3
📝 List your secrets

Make a short note of things to protect, like emails, phone numbers, or IDs, so it knows what to hide.

4
🛡️ Chat safely

As you talk to the AI, it swaps your secrets with safe stand-ins before sending, keeping them private.

5
🔧 Use local helpers

When running your own tools, everything switches back to real values so they work perfectly.

6
✨ See it restore

After the AI replies, your screen shows the full, natural text with secrets back in place.

🎉 Private AI power

Now you enjoy smart AI help without ever risking your personal details leaking out.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 19 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is opencode-vibeguard?

opencode-vibeguard is a JavaScript plugin for OpenCode that redacts sensitive strings like emails, phone numbers, IDs, UUIDs, IPs, and MAC addresses from chat messages before they hit LLM providers. It swaps them for irreversible placeholders using HMAC hashes tied to a session secret, then restores originals post-response for natural display and before local tool calls like bash or file edits. This keeps plaintext out of provider logs while ensuring tools run correctly locally.

Why is it gaining traction?

It stands out by handling full chat histories, tool inputs/outputs, and deep object redaction automatically, without disrupting OpenCode's streaming or session flow. Developers love the zero-config builtins for common secrets, plus easy custom patterns via a JSON config file loaded from project roots or env vars. npm install makes it drop-in ready, with debug logs that count replacements sans exposing secrets.

Who should use this?

OpenCode users building agents that read .env files, query databases, or execute shell commands with real credentials. Ideal for indie devs prototyping LLM tools locally who worry about accidental leaks to cloud providers, or teams auditing privacy in JavaScript-based AI workflows. Skip if you're not using OpenCode or tools.

Verdict

Grab it if OpenCode privacy gaps keep you up at night—it's a smart, lightweight fix despite 19 stars and 1.0% credibility signaling early maturity. Solid docs and tests make it low-risk to try, but watch for streaming glitches where placeholders flash briefly.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.