aj-geddes

Straylight AI — AI skill for Claude Code

AI community

Zero-knowledge credential proxy for AI agents.

How to install Straylight AI

This entry records only its repository, not the path inside it, so there is no exact command to give. Open aj-geddes/straylight-ai and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Straylight AI does

Zero-knowledge credential proxy for AI agents. Use AI, with Zero trust.

Alternatives in AI

  • Opencodex — Universal provider proxy for OpenAI Codex & Claude Code — use any LLM (Claude, Gemini, Grok, DeepSeek, Ollama… 12.3k ★
  • WindsurfAPI — Turn Windsurf / Devin Desktop's 100+ AI models (Claude, GPT, Gemini, DeepSeek, Kimi, GLM, SWE) into OpenAI-, A 3k ★
  • Phantom — An AI co-worker with its own computer 1.3k ★

README

Straylight-AI — Keep your API keys safe from AI coding assistants

Straylight-AI

Keep your API keys safe when using Claude Code, Cursor, and Windsurf.

Use AI, with Zero trust.

Build Status Version License: MIT Documentation

The Problem

Every time Claude Code reads your `.env` file, accesses shell history, or processes log output, your API keys enter its context window. From there they can be echoed in responses, logged to disk, or exfiltrated through prompt injection.

**This isn't theoretical.** CVE-2025-59536 demonstrated credential leakage via crafted API responses. CVE-2026-21852 showed API key exfiltration through malicious project configs.

The Solution

Straylight-AI is a self-hosted credential proxy that sits between your AI coding assistant and the outside world. You paste your API keys into a secure vault once. When Claude Code, Cursor, or Windsurf needs to call an API, Straylight injects the credential at the HTTP transport layer — **your keys never appear in the AI's context window, prompts, logs, or responses.**

Beyond API key injection, Straylight now covers database access, cloud provid