ifixai-aiifixai-aiOfficialยท1 Agent Skills Included

iFixAi

Audit your AI agent against business rules before failures happen

Audits AI agents against your business rules, org structure, and safety policies using adversarial test scenarios. Grades the agent with an independent judge model and returns an A-F scorecard showing exactly where it fails. Runs from the CLI, Claude Code, Codex, or any agent via a single install command, with results in under two minutes.
npx skills add ifixai-ai/iFixAi --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install iFixAi?โ–ผ

Run `npx skills add ifixai-ai/iFixAi --all -g -y` in your terminal to install the full suite globally.

What does iFixAi actually test?โ–ผ

It checks whether your agent follows your business rules, permission boundaries, and safety policies by probing it with adversarial scenarios, then grades the results.

Can iFixAi test my deployed agent?โ–ผ

Yes. Point it at your agent's live HTTP endpoint to test the real deployed system, or fall back to testing the bare model underneath.

Does iFixAi work with Claude Code and Codex?โ–ผ

Yes. It installs as a plugin in Claude Code and Codex, or as a skill command in Cursor, VS Code, Windsurf, Cline, Continue, Gemini, and Zed.

Do I need coding skills to run an audit?โ–ผ

No. The guided wizard or your AI agent walks you through setup, builds the test fixture, and explains the scorecard in plain English.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’