Launching today

Agent Security

Launching today

Decide API before side effects: allow, deny, or pending

1 follower

Decide API before side effects: allow, deny, or pending

1 follower

Visit website

Agents plan tool calls. Agent Security sits between the runtime and integrations (email, tickets, APIs). Each call hits the policy engine (decide API) before side effects run: allow, deny, or pending. Allow executes the tool. Deny blocks it. Pending goes to a human-in-the-loop approval queue. Every decision is recorded on an append-only audit trail (workspace events, export later).

Free Options

Launch tags:Developer Tools•Artificial Intelligence•Security

Launch Team / Built With

Viktor.comAn AI coworker that actually does the work

Promoted

Maker

📌

I built Agent Security because logging after an agent already sent the email is too late. The gap I kept hitting: nothing sits between “the model wants a tool” and “the integration actually ran.” So the product is that layer. The runtime plans the tool call. A policy engine (decide API) evaluates it before side effects: allow, deny, or pending. Allow executes (email, tickets, APIs). Deny blocks. Pending goes to a human-in-the-loop queue. Decisions land on an append-only audit trail. Happy to walk through the runtime flow if useful.

Report

1d ago

Forum Threads

p/agent-security

•

1d ago

Agent Security — decide before the tool runs

Agent Security is a policy layer between the agent runtime and integrations (email, tickets, APIs).

The runtime plans a tool call. The decide API evaluates it before side effects: allow, deny, or pending.
Allow executes. Deny blocks.
Pending goes to a human-in-the-loop queue. Decisions land on an append-only audit trail.

Not a prompt wrapper. A gate on the tool path.

Live page: https://alfanestlabs.com/agent-s...

View all