Back to directory

Harden.Run AIF

Harden.Run AIF is a local safety layer for coding agents that installs native hooks in tools such as Claude Code, Codex, Cursor and Gemini CLI to block risky actions before they run, with a local dashboard and anonymous, code-free operational telemetry.

Harden.Run AIF

Harden.Run AIF, short for Agentic Integrity Foundation, is a local system for coding-agent safety. It installs through a shell script and is configured, demonstrated and inspected through a small set of commands, including a terminal dashboard view.

It integrates with several agents at different depths: native hooks with pre-execution blocking for Claude Code, Codex, Cursor and Gemini CLI, hooks for Kiro's main agent with a proxy for subagent MCP calls, and a bridge process that blocks and steers OpenClaw and Hermes. Telemetry is opt-out and limited to anonymous operational counters such as version, OS and adapter usage; it explicitly never collects code, commands, file paths, prompts or secrets, and a command prints the exact payload before it is sent.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.