GitHub Repository

Zero-trust repo intake for AI coding agents — scan the instruction environment before Claude Code, Cursor, Codex, or Gemini touches a repo. Ships its own false-negative ledger.

3 starsPython

Agent-zero-trust – scan a repo before your AI coding agent reads it

Name: Agent-zero-trust – scan a repo before your AI coding agent reads it
Availability: InStock
Author: ralfyishere

by ralfyishere·Jul 8, 2026·1 point·1 comment

Visit Project View on HN

AI Analysis

●●●BangerBig BrainSolve My ProblemShip It

Stdlib-only scanner blocks prompt injections before Cursor or Claude Code ever sees them.

Strengths

•Treats repos as instruction environments, catching hidden HTML comments and config overrides.
•Stdlib-only Python design prevents supply chain attacks within the scanner itself.
•CI-ready exit codes and JSON output enable automated gating before agent access.

Weaknesses

•Relies on pattern matching, may miss novel semantic injection techniques without LLM assistance.
•Niche utility until AI agents become the default way everyone interacts with codebases.