# Harden AIF

Harden AIF checks supported coding-agent tool calls before execution. It applies local policy decisions such as allowing, blocking, redacting, or asking for confirmation.

Canonical URL: https://aiidelist.com/ide/harden-aif

Language: en

Updated: 2026-09-11

## Overview

- Category: Developer Workflow Tools
- Local policy enforcement for tool calls made through supported coding-agent integrations.
- Editor base: CLI
- Platforms: macOS, Linux
- Open source: No

## Quick verdict

A focused option for developers who want to inspect agent actions before execution and have compatible hardware.

## Best for

- Developers adding local controls to supported coding agents

## Strengths

- Adds decisions at the point where supported agent tools are invoked.
- Processes security decisions locally rather than requiring a cloud account.

## Limitations

- Windows-only environments
- Users expecting model acceleration on every daemon-supported machine
- Full model operation requires Apple Silicon; published guidance calls for at least 16 GB RAM and 15 GB disk space.
- Coverage depends on the configured integration; it is not a universal operating-system sandbox.

## Inspect the proposed action before it runs

Harden AIF sits between a supported agent integration and tool execution. A policy can allow a routine operation while stopping or escalating an operation that needs review. Its decisions are distinct from the filesystem and network boundaries provided by a sandbox.

A useful evaluation separates harmless permitted actions from harmless actions that your test policy intentionally forbids. That makes it possible to check both outcomes without using real secrets or destructive commands.

## Check one integration end to end

1. Compare your machine with the requirements on the [Harden website](https://harden.run/), including the distinction between daemon and full-model support.
2. Install using the official instructions and run the configuration flow for one supported coding agent.
3. Start a disposable project and issue a simple read-only task, then inspect the local decision record.
4. Test a deliberately restricted but harmless action and confirm that the configured block or approval behavior occurs.

After installation, configuration starts with:

```bash
aif configure
```

## Evaluate coverage as well as interruptions

Review which tools the integration actually intercepts and whether legitimate development tasks are repeatedly interrupted. Keep your existing workspace permissions in place while measuring that behavior. The relevant result is understandable enforcement on your workflow, not simply the number of blocked calls.

## Features

### Execution policy

- Checks supported tool calls before execution
- Allow, block, redact, ask, and log outcomes
- Local decision history

### Agent and hardware support

- Integrations include Claude Code, Codex, Cursor, Antigravity CLI, and Kiro
- Full local model path uses Apple Silicon and Metal
- CLI and daemon support macOS and Linux x86-64

## Pricing

free individual protection

- Individual protection: $0 — Core local protection is free for individuals. No account is required; suitable local hardware is still needed.

Pricing checked: 2026-09-12

## Privacy and data handling

Security decisions run locally. Optional telemetry can be disabled. Connected coding agents retain their own model-provider data handling.

## Sources

- [Official website](https://harden.run/)
- [Installation](https://harden.run/)
- [Pricing](https://harden.run/)
- [Harden product, installation, and hardware requirements](https://harden.run/)

Last checked: 2026-09-12