cd /news/ai-agents/we-didn-t-feel-comfortable-giving-ou… · home › topics › ai-agents › article
[ARTICLE · art-172888] src=dev.to ↗ pub= topic=ai-agents verified=true sentiment=↑ positive

We didn't feel comfortable giving our AI agent our API keys

Two brothers who run Hermes-based AI agents built aiSecurityWarden, a human approval gate that sits between agents and their credentials, after growing uncomfortable pasting live Gmail app passwords, GitHub tokens and model API keys into config files. The hosted service uses zero-knowledge storage, 2FA-protected approvals and deny-by-default gating on connectors that touch secrets, sending an approval request to the user's phone with the actual action body before an agent can send, write or spend. The developers argue that system-prompt warnings, in-agent approval prompts and scoped short-lived keys all fail against prompt injection because the compromised agent is the one making the request.

recommendation

Route credential-touching agent actions through an external approval gate that shows the actual request body, and gate only sends, writes and money rather than everything.

more in playbook: Securing agents and AI tools
by read3 min views3 publishedNov 19, 2026

My brother and I have been running AI agents for a few months now — Hermes-based agents that fetch things, send emails, touch GitHub, do real work while we do other things. Once an agent actually finishes a real task for you, it's hard to go back.

Then comes the part nobody talks about.

The agent needs to send Gmail: here's an app password. It needs GitHub: here's a personal access token. It needs to call the model: here's an API key. One by one we pasted live credentials into config files and hoped that nothing weird would ever show up in the context window.

We kept asking ourselves the same question: what happens the day our agent gets lied to?

While we were debating, the industry was having a bad semester: Four very different worlds — frontier labs, AI safety researchers, cloud deployments, enterprises. One pattern:

The agent already had the keys.

The obvious answer is "tell the agent not to leak the keys." Add a line to the system prompt: never reveal your credentials. Cute. Prompt injection is precisely the attack that edits what your agent believes. If one email can change your agent's behavior, it can ignore one policy sentence.

The second obvious answer is an approval prompt inside the agent's own UI. But the hijacked agent is the one writing the request you approve. Asking a compromised process to explain what it's about to do is not a security boundary — it's a suggestion box.

"Use scoped, short-lived keys" is the third answer. Real improvement, still insufficient: scope limits what a stolen key can do, it doesn't stop the theft. An OAuth token that can send one email is an attacker who can send email.

That's the whole idea behind aiSecurityWarden. My brother and I built it for ourselves first — we were running Hermes and genuinely did not want to hand it our Gmail, Github or other credentials. It's a human approval gate that sits between your agents and their secrets:

In practice: your agent tries to send an email, your phone buzzes, you read the actual body and tap Approve — or Deny with a comment the agent has to respect. Deny-by-default on gated connectors; everything that doesn't touch a secret keeps running free.

And there's nothing hobby-project about how it's built. Zero-knowledge storage, 2FA-protected approvals, deny-by-default — the same guarantees an enterprise security team would demand, just without the enterprise price tag, the procurement process, or the sales call. You shouldn't have to be a company to deserve real security infrastructure for your agents.

Friction. If you gate everything you'll hate yourself by day two — gate the handful of actions that actually matter (sends, writes, money) and let the rest run. It's also our hosted service today, not a self-hosted tarball; there's a free option at aisecuritywarden.com.

We didn't invent the idea — "don't hand the executor your keys" is the oldest rule in security. What changed is that the executor is now a very convincing chatbot that reads untrusted text all day long. And I'd rather not admit it, but the news cycle keeps validating the product for us.

If you run agents with real credentials, I'd genuinely like your feedback — especially where you think we're wrong. What deserves gating? How much context should an approval card show before a human can actually judge it? Where does human-in-the-loop break a real workflow? Tell me in the comments, or at aisecuritywarden.com. The approval flow in 90 seconds:

── more in #ai-agents 4 stories · sorted by recency
── more on @aisecuritywarden 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/we-didn-t-feel-comfo…] indexed:0 read:3min 2026-11-19 · —