Survey: Does your AI agent have an identity problem?

[Marsssssssssssdsss]

Hi everyone 👋

I'm reaching out to the AI agent community because I think we're all heading toward the same problem, and I'd love your input.

The question we're exploring:

Every SaaS platform has user accounts — for humans. But when your AI agent calls an API, it borrows a human's credentials. The AI itself has no identity.

If you're building AI agents that call external services (Shopify, Slack, Notion, GitHub, etc.) — how are you handling authentication across multiple platforms right now?

We've put together a short 3-minute survey to collect experiences from people who've hit this wall (or think they will). No sign-up, no product pitch — just honest questions to understand how the community is handling this:

🔗 Take the survey →

I'd really appreciate if you could spare a few minutes to share your thoughts. I'll share the results openly once we have enough data.

Thank you! 🙏

[mitalidaduria]

Dealing with this right now for a Multi-Agent LangGraph project. If you just pass raw human credentials, you completely lose track of data lineage, which is a massive bottleneck for enterprise compliance.
Currently, we're handling this by using a centralized credential manager that injects scoped, short-lived OAuth tokens into the agent's runtime context on a per-tool basis. It works, but it's duct-tape. We really need a formal Agent Catalog registry in data governance platforms to manage these machine identities properly.