iFixAi
Independent Auditing of AI Agents. Run by human or the agent itself, to answer the most crucial question in the AI Agent Economy. Is the agent doing what is sup
What is iFixAi?
Independent Auditing of AI Agents. Run by human or the agent itself, to answer the most crucial question in the AI Agent Economy. Is the agent doing what is supposed to do? With iFixAi you can have this answer in less than 120 seconds.
Overview of iFixAi
What it is
iFixAi is an independent auditing tool for AI agents. It can be run by a human or by the agent itself, and provides an answer to the question of whether an agent is doing what it is supposed to do. The tool evaluates agents based on business KPIs and organizational structure, striking a balance between AI Red Teaming and Operational Assurance. Results are delivered in less than 120 seconds.
Key features
- Runs a unified diagnostic through a CLI guided wizard, explicit flags, or as a plugin/skill inside an agent - Supports multiple providers and judges, including the option for independent cross-vendor grading - Generates JSON and Markdown reports with a rich terminal scorecard - Provides pre-built test suites (smoke, strategic, core, extended, all) selectable via wizard or flag - Includes a built-in mock fixture with seeded defects for testing the pipeline without API keys
Use cases
- Testing an AI agent's adherence to business KPIs before deployment - Auditing agent performance in CI/CD pipelines using automated scripts - Running adversarial diagnostics alongside operational assurance checks - Getting a citable grade from an independent provider grading the agent
Getting started
Guided wizard (recommended): bash pip install ifixai[openai] ifixai setup ifixai run Plugin for Claude Code: /plugin marketplace add ifixai-ai/iFixAi /plugin install ifixai@ifixai-community Skill for any agent: bash uvx ifixai install --agents cursor Explicit flags: bash pip install ifixai[anthropic] ifixai run --provider mock --api-key not-used --eval-mode self
License and pricing
The license is Apache-2.0. Pricing is not stated; the project is categorized as open source.
Based on the project's public README and repository metadata.