Diagnostic

Is Your AI Agent Actually Working?

Answer 10 questions. Get a score. Fix what's broken.

1. Do you have to talk to your AI for it to start working?

Yes, I have to prompt it every time (Reactive)

No, it runs on a schedule autonomously (Proactive)

2. Does your agent remember what you told it 3 days ago?

No, every session is a clean slate

Yes, it has persistent memory architecture

3. Where does your agent live?

In a browser tab or web app

On a dedicated VPS / local server

4. How is it secured?

Generic browser sandboxing

Defined "Gateway" with restricted tool access

5. Are you optimizing model costs?

No, I use the most expensive model for everything

Yes, I route simple tasks to cheaper models

6. How does it handle long-term tasks?

I have to stay in the chat to keep it running

It runs "Heartbeat" loops independently

7. Does it manage other agents?

No, it does everything in one context

Yes, it orchestrates specialized sub-agents

8. How are its instructions defined?

One large "system prompt"

Structured SOUL.md + Skill files

9. Can it navigate the web?

No, it's limited to text input

Yes, via browser automation skills

10. Do you have a log of its autonomous work?

No, I just see the final output

Yes, detailed daily build logs

0 Health Score

Needs Assessment

Critical Fix List

Get Your Detailed roadmap

We'll send you a step-by-step roadmap to reach a 100/100 score.

Join the Marketplace for a 100/100 Score →