Stop letting Claude Code burn your tokens and leak API keys
Local firewall for AI agents stops token burn before requests leave your machine.
Stop AI coding agents from leaking your API keys. Local proxy + MCP that swaps real secrets for phm_ tokens — works with Claude Code, Cursor, Windsurf, and Codex.
Proxy tokens worthless if leaked, real keys never enter LLM context windows.
Developers using AI coding agents like Cursor, Claude Code, or Copilot
Doppler · git-secrets · Vault
Local firewall for AI agents stops token burn before requests leave your machine.
MITM proxy swaps fake keys for real ones so agents never see credentials.
Drop-in Claude API proxy with real-time cost dashboard—but Anthropic's own billing UI exists.
Agents manage their own context window through a transparent local proxy.
Yet another secret scanner, but this one's a single Python file.
Credential vaulting proxy for OpenClaw, but solves a narrow ecosystem problem.