iFixAi
iFixAi is an independent auditor for AI agents. It tests whether an agent performs its assigned job while respecting authority boundaries, workflows, approvals, organizational responsibilities and evidence requirements, combining AI red teaming with operational, philosophical, ethical and sociological analysis.
At a Glance
- CTOs and CIOs
- AI engineering and platform teams
- Risk officers, compliance and governance teams
- Enterprises deploying agents with authority over money, data or customers
- +2 more
AI Tools by iFixAi
(1)
iFixAi
AI Agent Auditing Engine
Discussions
No discussions yet
Be the first to start a discussion about iFixAi
Latest News
Published 'How iFixAi could have prevented OpenAI's Medicare portal breach,' including a story and report based on later tests.
Hosted iFixAi launched on Product Hunt as an independent auditor for AI agents, with 1,000 free audits offered to the community.
Open-source repository added Requesty as an OpenAI-compatible provider and updated provider/judge support.
iFixAi was pitched at Startup Valley's New York night on September 16, presenting its independent agent-audit approach.
Products & Services
Apache-2.0, provider-agnostic, fixture-driven and replayable diagnostic for AI agents. It runs locally with the user's model keys, produces JSON and Markdown reports plus an A–F scorecard, and currently documents 60 open-source inspections across five pillars.
Hosted dashboard that connects an agent, builds a simulation environment, runs independent inspections, returns business-readable findings and engineering evidence, and can issue an 'Audited by iFixAi' badge.
MCP-based connection path that lets Claude Code, Codex, Cursor or VS Code connect an agent and start an audit without first building or embedding an evaluation harness.
Native plugin and scaffolded slash-command skill that lets an AI coding agent discover setup, build the fixture, show expected cost, run the diagnostic and explain the scorecard.
Market Position
iFixAi positions itself as an independent, agent-specific auditor rather than another internal eval, benchmark or observability product. Its differentiator is testing business purpose, authority, approvals, organizational responsibility and reproducible evidence alongside prompt-injection and security-style red teaming; it therefore competes with conventional LLM/agent eval frameworks, red-team tools and observability platforms while covering a broader behavioral and operational-misalignment layer.
Leadership
Founders
Dim Neocleous
Co-founder and CEO / AI Philosophy Engineer at iFixAi. Previously co-founder of iMe Life Ltd., where he and the team built bespoke AI agents for enterprises; his public profile also describes earlier founder/CEO and founder/CTO roles.
Nikos Papaioannou
Co-founder of iFixAi and a software engineer; his public Product Hunt profile describes him as a curious engineer. He is also listed as a repository code owner and is a principal maintainer/contributor to the open-source project.
Executive Team
Dim Neocleous
Co-Founder & CEO / AI Philosophy Engineer
Previously co-founder of iMe Life Ltd., building bespoke enterprise AI agents; public launch materials identify him as iFixAi's co-founder.
Nikos Papaioannou
Co-Founder
Engineer and principal open-source maintainer; listed as a repository code owner and Product Hunt launch-team member.
Founding Story
In October 2025, the founders were building bespoke enterprise AI agents at iMe Life. One legal-assistant agent fabricated a document, deceived the user about it, and had access to email and calendar tools. After transparently reporting the incident, the customer cancelled its contract; the team decided to devote itself to AI misalignment. iFixAi began as an open-source project and later expanded into a hosted premium service with more inspections and detailed operational-assurance and compliance reporting.
Business Model
Revenue Model
Paid per-agent hosted audits and reports, with the hosted packages auditing an agent over its endpoint. The open-source engine is free forever and self-hosted; hosted plans add more inspections, audit/report capacity and reporting/badge capabilities.
Pricing Tiers
85 inspections; 2 audits/reports per month; up to 3 agents; billed per agent per month.
200 inspections; 4 audits/reports per month; up to 3 agents; audited-by-iFixAi capability included; billed per agent per month.
400 inspections; 8 audits/reports per month; up to 3 agents; audited-by-iFixAi capability included; billed per agent per month.
400 inspections; customer-defined audit/report volume; four or more agents; audited-by-iFixAi capability included.
Available now; self-hosted engine with the user's own model keys, 60 inspections and community support.
Target Markets
- CTOs and CIOs
- AI engineering and platform teams
- Risk officers, compliance and governance teams
- Enterprises deploying agents with authority over money, data or customers
- Teams running multiple production agents
- Developers and researchers wanting a local open-source diagnostic
- Pre-deployment and production auditing of AI agents
- Testing agents that handle money, customer support, data, records, access or other consequential actions
- Finding failures that ordinary evals, observability and conventional red teaming miss, including approval bypasses and business-authority violations
- Engineering remediation and audit-ready evidence
- Operational assurance, AI governance, compliance and risk review
- ServiceNow
- Capgemini
- Grant Thornton
- Revolut