[GH-ISSUE #498] Add BlackVault — Proxy gateway for AI API keys #14208

New Issue

GiteaMirror · 2026-07-12T22:15:28-05:00

GiteaMirror commented

2026-07-12 22:15:28 -05:00

Originally created by @venkat22022202 on GitHub (Feb 13, 2026).
Original GitHub issue: https://github.com/Shubhamsaboo/awesome-llm-apps/issues/498

Project Submission

Project Name: BlackVault
URL: https://github.com/venkat22022202/black-vault
Live Demo: https://black-vault-murex.vercel.app
License: MIT
Category: AI Tools / Agent Infrastructure

Description

Open-source proxy gateway for AI API keys. Instead of sharing real API keys with AI agents, you generate proxy tokens. BlackVault injects the real key server-side and forwards requests to OpenAI, Anthropic, or Google AI with full SSE streaming. Kill a session = instant access revocation. No key rotation needed.

Key Features

AES-256-GCM Encrypted Vault — API keys are encrypted at rest
Per-Session Cost Tracking — monitor spend per agent session
Instant Kill Switch — revoke agent access immediately, no key rotation
Agent Registry — manage proxy tokens for multiple agents
Workflow Blueprints — pre-built agent workflow templates
Full SSE Streaming — supports OpenAI, Anthropic, and Google AI providers

Tech Stack

Next.js 16, tRPC, Neon Postgres, Upstash Redis, Clerk Auth

Why This Belongs Here

BlackVault solves a critical problem in the LLM ecosystem: securely sharing API keys with AI agents without exposing the real credentials. It's a practical, open-source tool that any developer building with LLMs can use to protect their API keys and control agent access.

Originally created by @venkat22022202 on GitHub (Feb 13, 2026). Original GitHub issue: https://github.com/Shubhamsaboo/awesome-llm-apps/issues/498 ## Project Submission **Project Name:** BlackVault **URL:** https://github.com/venkat22022202/black-vault **Live Demo:** https://black-vault-murex.vercel.app **License:** MIT **Category:** AI Tools / Agent Infrastructure ## Description Open-source proxy gateway for AI API keys. Instead of sharing real API keys with AI agents, you generate proxy tokens. BlackVault injects the real key server-side and forwards requests to OpenAI, Anthropic, or Google AI with full SSE streaming. Kill a session = instant access revocation. No key rotation needed. ## Key Features - **AES-256-GCM Encrypted Vault** — API keys are encrypted at rest - **Per-Session Cost Tracking** — monitor spend per agent session - **Instant Kill Switch** — revoke agent access immediately, no key rotation - **Agent Registry** — manage proxy tokens for multiple agents - **Workflow Blueprints** — pre-built agent workflow templates - **Full SSE Streaming** — supports OpenAI, Anthropic, and Google AI providers ## Tech Stack Next.js 16, tRPC, Neon Postgres, Upstash Redis, Clerk Auth ## Why This Belongs Here BlackVault solves a critical problem in the LLM ecosystem: securely sharing API keys with AI agents without exposing the real credentials. It's a practical, open-source tool that any developer building with LLMs can use to protect their API keys and control agent access.

GiteaMirror closed this issue

2026-07-12 22:15:29 -05:00

GiteaMirror commented

2026-07-12 22:15:31 -05:00

@awesomekoder commented on GitHub (Feb 15, 2026):

Hey @venkat22022202 — thanks for sharing BlackVault! However, this repo focuses on self-contained AI/LLM app tutorials, not tool/infrastructure listings. This reads more like a project promotion than a tutorial request. If you'd like to contribute, consider submitting a PR with a self-contained tutorial that teaches a concept (e.g., "Build a secure AI API proxy with rate limiting") using your approach.

@awesomekoder commented on GitHub (Feb 15, 2026): Hey @venkat22022202 — thanks for sharing BlackVault! However, this repo focuses on self-contained AI/LLM app tutorials, not tool/infrastructure listings. This reads more like a project promotion than a tutorial request. If you'd like to contribute, consider submitting a PR with a self-contained tutorial that teaches a concept (e.g., "Build a secure AI API proxy with rate limiting") using your approach.

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/awesome-llm-apps#14208